{"id":13570262,"url":"https://github.com/satellite-image-deep-learning/techniques","last_synced_at":"2025-05-11T11:02:29.631Z","repository":{"id":37318903,"uuid":"129712017","full_name":"satellite-image-deep-learning/techniques","owner":"satellite-image-deep-learning","description":"Techniques for deep learning with satellite \u0026 aerial imagery","archived":false,"fork":false,"pushed_at":"2025-05-07T16:42:55.000Z","size":29324,"stargazers_count":9321,"open_issues_count":4,"forks_count":1563,"subscribers_count":280,"default_branch":"master","last_synced_at":"2025-05-07T16:49:40.158Z","etag":null,"topics":["convolutional-neural-networks","dataset","datasets","deep-learning","deep-neural-networks","earth-observation","image-classification","keras","machine-learning","object-detection","python","pytorch","remote-sensing","satellite-data","satellite-imagery","satellite-images","sentinel","tensorflow"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/satellite-image-deep-learning.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":".github/FUNDING.yml","license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null},"funding":{"github":"robmarkcole"}},"created_at":"2018-04-16T08:42:09.000Z","updated_at":"2025-05-07T16:42:59.000Z","dependencies_parsed_at":"2024-03-27T06:28:17.402Z","dependency_job_id":"602e3ca2-9d09-417f-aea8-076e1ed71ff6","html_url":"https://github.com/satellite-image-deep-learning/techniques","commit_stats":{"total_commits":1292,"total_committers":24,"mean_commits":"53.833333333333336","dds":0.04566563467492257,"last_synced_commit":"bb676caeac5b8f59ad92d2281084a624a41f7d0b"},"previous_names":["robmarkcole/satellite-image-deep-learning"],"tags_count":22,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/satellite-image-deep-learning%2Ftechniques","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/satellite-image-deep-learning%2Ftechniques/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/satellite-image-deep-learning%2Ftechniques/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/satellite-image-deep-learning%2Ftechniques/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/satellite-image-deep-learning","download_url":"https://codeload.github.com/satellite-image-deep-learning/techniques/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253554071,"owners_count":21926610,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["convolutional-neural-networks","dataset","datasets","deep-learning","deep-neural-networks","earth-observation","image-classification","keras","machine-learning","object-detection","python","pytorch","remote-sensing","satellite-data","satellite-imagery","satellite-images","sentinel","tensorflow"],"created_at":"2024-08-01T14:00:50.328Z","updated_at":"2025-05-11T11:02:29.414Z","avatar_url":"https://github.com/satellite-image-deep-learning.png","language":null,"readme":"\u003cdiv align=\"center\"\u003e\n  \u003cp\u003e\n    \u003ca href=\"https://www.satellite-image-deep-learning.com/\"\u003e\n        \u003cimg src=\"images/logo.png\" width=\"700\"\u003e\n    \u003c/a\u003e\n\u003c/p\u003e\n\n# 👉 [satellite-image-deep-learning.com](https://www.satellite-image-deep-learning.com/) 👈\n\n\u003c/div\u003e\n\n## Introduction\n\nDeep learning has revolutionized the analysis and interpretation of satellite and aerial imagery, addressing unique challenges such as vast image sizes and a wide array of object classes. This repository provides an exhaustive overview of deep learning techniques specifically tailored for satellite and aerial image processing. It covers a range of architectures, models, and algorithms suited for key tasks like classification, segmentation, and object detection.\n\n**How to use this repository:** use `Command + F` (Mac) or `CTRL + F` (Windows) to search this page for e.g. 'SAM'\n\n## Techniques\n\n- [Classification](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#classification)\n- [Segmentation](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#segmentation)\n- [Object detection](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#object-detection)\n- [Regression](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#regression)\n- [Cloud detection \u0026 removal](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#cloud-detection--removal)\n- [Change detection](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#change-detection)\n- [Time series](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#time-series)\n- [Crop classification](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#crop-classification)\n- [Crop yield \u0026 vegetation forecasting](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#crop-yield--vegetation-forecasting)\n- [Generative networks](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#generative-networks)\n- [Autoencoders, dimensionality reduction, image embeddings \u0026 similarity search](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#autoencoders-dimensionality-reduction-image-embeddings--similarity-search)\n- [Few \u0026 zero shot learning](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#few--zero-shot-learning)\n- [Self-supervised, unsupervised \u0026 contrastive learning](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#self-supervised-unsupervised--contrastive-learning)\n- [SAR](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#sar)\n- [Large vision \u0026 language models (LLMs \u0026 LVMs)](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#large-vision--language-models-llms--lvms)\n- [Foundational models](https://github.com/satellite-image-deep-learning/techniques?tab=readme-ov-file#foundational-models)\n\n#\n## Classification\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/merced.png\" width=\"600\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eThe UC merced dataset is a well known classification dataset.\u003c/b\u003e\n\u003c/p\u003e\n\nClassification is a fundamental task in remote sensing data analysis, where the goal is to assign a semantic label to each image, such as 'urban', 'forest', 'agricultural land', etc. The process of assigning labels to an image is known as image-level classification. However, in some cases, a single image might contain multiple different land cover types, such as a forest with a river running through it, or a city with both residential and commercial areas. In these cases, image-level classification becomes more complex and involves assigning multiple labels to a single image. This can be accomplished using a combination of feature extraction and machine learning algorithms to accurately identify the different land cover types. It is important to note that image-level classification should not be confused with pixel-level classification, also known as semantic segmentation. While image-level classification assigns a single label to an entire image, semantic segmentation assigns a label to each individual pixel in an image, resulting in a highly detailed and accurate representation of the land cover types in an image. Read [A brief introduction to satellite image classification with neural networks](https://medium.com/@robmarkcole/a-brief-introduction-to-satellite-image-classification-with-neural-networks-3ce28be15683)\n\n- Land classification on Sentinel 2 data using a [simple sklearn cluster algorithm](https://github.com/acgeospatial/Satellite_Imagery_Python/blob/master/Clustering_KMeans-Sentinel2.ipynb)\n  \n- [Multi-Label Classification of Satellite Photos of the Amazon Rainforest using keras](https://machinelearningmastery.com/how-to-develop-a-convolutional-neural-network-to-classify-satellite-photos-of-the-amazon-rainforest/)\n\n- [EuroSat-Satellite-CNN-and-ResNet](https://github.com/Rumeysakeskin/EuroSat-Satellite-CNN-and-ResNet) -\u003e Classifying custom image datasets by creating Convolutional Neural Networks and Residual Networks from scratch with PyTorch\n\n- [Detecting Informal Settlements from Satellite Imagery using fine-tuning of ResNet-50 classifier](https://blog.goodaudience.com/detecting-informal-settlements-using-satellite-imagery-and-convolutional-neural-networks-d571a819bf44) with [repo](https://github.com/dymaxionlabs/ap-latam)\n\n-  [Land-Cover-Classification-using-Sentinel-2-Dataset](https://github.com/raoofnaushad/Land-Cover-Classification-using-Sentinel-2-Dataset) -\u003e [well written Medium article](https://raoofnaushad7.medium.com/applying-deep-learning-on-satellite-imagery-classification-5f2588b932c1) accompanying this repo but using the EuroSAT dataset\n\n- [Slums mapping from pretrained CNN network](https://github.com/deepankverma/slums_detection) on VHR (Pleiades: 0.5m) and MR (Sentinel: 10m) imagery\n\n- [Comparing urban environments using satellite imagery and convolutional neural networks](https://github.com/adrianalbert/urban-environments) -\u003e includes interesting study of the image embedding features extracted for each image on the Urban Atlas dataset\n\n- [RSI-CB](https://github.com/lehaifeng/RSI-CB) -\u003e A Large Scale Remote Sensing Image Classification Benchmark via Crowdsource Data. See also [Remote-sensing-image-classification](https://github.com/aashishrai3799/Remote-sensing-image-classification)\n\n- [NAIP_PoolDetection](https://github.com/annaptasznik/NAIP_PoolDetection) -\u003e modelled as an object recognition problem, a CNN is used to identify images as being swimming pools or something else - specifically a street, rooftop, or lawn\n\n- [Land Use and Land Cover Classification using a ResNet Deep Learning Architecture](https://www.luigiselmi.eu/eo/lulc-classification-deeplearning.html) -\u003e uses fastai and the EuroSAT dataset\n\n- [Vision Transformers Use Case: Satellite Image Classification without CNNs](https://medium.com/nerd-for-tech/vision-transformers-use-case-satellite-image-classification-without-cnns-2c4dbeb06f87)\n\n- [WaterNet](https://github.com/treigerm/WaterNet) -\u003e a CNN that identifies water in satellite images\n\n- [Road-Network-Classification](https://github.com/ualsg/Road-Network-Classification) -\u003e Road network classification model using ResNet-34, road classes organic, gridiron, radial and no pattern\n\n- [Scaling AI to map every school on the planet](https://developmentseed.org/blog/2021-03-18-ai-enabling-school-mapping)\n\n- [satellite-crosswalk-classification](https://github.com/rodrigoberriel/satellite-crosswalk-classification)\n\n- [Implementation of the 3D-CNN model for land cover classification](https://medium.com/geekculture/remote-sensing-deep-learning-for-land-cover-classification-of-satellite-imagery-using-python-6a7b4c4f570f) -\u003e uses the Sundarbans dataset, with [repo](https://github.com/syamkakarla98/Satellite_Imagery_Analysis)\n\n- [SSTN](https://github.com/zilongzhong/SSTN) -\u003e Spectral-Spatial Transformer Network for Hyperspectral Image Classification: A FAS Framework\n\n- [SatellitePollutionCNN](https://github.com/arnavbansal1/SatellitePollutionCNN) -\u003e A novel algorithm to predict air pollution levels with state-of-art accuracy using deep learning and GoogleMaps satellite images\n\n- [PropertyClassification](https://github.com/Sardhendu/PropertyClassification) -\u003e Classifying the type of property given Real Estate, satellite and Street view Images\n\n- [remote-sense-quickstart](https://github.com/CarryHJR/remote-sense-quickstart) -\u003e classification on a number of datasets, including with attention visualization\n\n- [Satellite image classification using multiple machine learning algorithms](https://github.com/tanmay-delhikar/satellite-image-analysis-ml)\n\n- [satsense](https://github.com/DynaSlum/satsense) -\u003e land use/cover classification using classical features including HoG \u0026 NDVI\n\n- [PyTorch_UCMerced_LandUse](https://github.com/GeneralLi95/PyTorch_UCMerced_LandUse)\n\n- [EuroSAT-image-classification](https://github.com/artemisart/EuroSAT-image-classification)\n\n- [landcover_classification](https://github.com/reidfalconer/landcover_classification) -\u003e using fast.ai on EuroSAT\n\n- [IGARSS2020_BWMS](https://github.com/jiankang1991/IGARSS2020_BWMS) -\u003e Band-Wise Multi-Scale CNN Architecture for Remote Sensing Image Scene Classification with a novel CNN architecture for the feature embedding of high-dimensional RS images\n\n- [image.classification.on.EuroSAT](https://github.com/canturan10/image.classification.on.EuroSAT) -\u003e solution in pure pytorch\n\n- [hurricane_damage](https://github.com/allankapoor/hurricane_damage) -\u003e Post-hurricane structure damage assessment based on aerial imagery\n\n- [openai-drivendata-challenge](https://github.com/buildwithcycy/openai-drivendata-challenge) -\u003e Using deep learning to classify the building material of rooftops (aerial imagery from South America)\n\n- [is-it-abandoned](https://github.com/zach-brown-18/is-it-abandoned) -\u003e Can we tell if a house is abandoned based on aerial LIDAR imagery?\n\n- [BoulderAreaDetector](https://github.com/pszemraj/BoulderAreaDetector) -\u003e CNN to classify whether a satellite image shows an area would be a good rock climbing spot or not\n\n- [ISPRS_S2FL](https://github.com/danfenghong/ISPRS_S2FL) -\u003e Multimodal Remote Sensing Benchmark Datasets for Land Cover Classification with A Shared and Specific Feature Learning Model\n\n- [Brazilian-Coffee-Detection](https://github.com/MrSquidward/Brazilian-Coffee-Detection) -\u003e uses Keras with public dataset\n\n- [tf-crash-severity](https://github.com/SoySauceNZ/tf-crash-severity) -\u003e predict the crash severity for given road features contained within satellite images\n\n- [ensemble_LCLU](https://github.com/burakekim/ensemble_LCLU) -\u003e Deep neural network ensembles for remote sensing land cover and land use classification\n\n- [cerraNet](https://github.com/MirandaMat/cerraNet-v2) -\u003e contextually classify the types of use and coverage in the Brazilian Cerrado\n\n- [Urban-Analysis-Using-Satellite-Imagery](https://github.com/mominali12/Urban-Analysis-Using-Satellite-Imagery) -\u003e classify urban area as planned or unplanned using a combination of segmentation and classification\n\n- [ChipClassification](https://github.com/yurithefury/ChipClassification) -\u003e Deep learning for multi-modal classification of cloud, shadow and land cover scenes in PlanetScope and Sentinel-2 imagery\n\n- [DeeplearningClassficationLandsat-tImages](https://github.com/VinayarajPoliyapram/DeeplearningClassficationLandsat-tImages) -\u003e Water/Ice/Land Classification Using Large-Scale Medium Resolution Landsat Satellite Images\n\n- [wildfire-detection-from-satellite-images-ml](https://github.com/shrey24/wildfire-detection-from-satellite-images-ml) -\u003e detect whether an image contains a wildfire, with example flask web app\n\n- [mining-discovery-with-deep-learning](https://github.com/remis/mining-discovery-with-deep-learning) -\u003e Mining and Tailings Dam Detection in Satellite Imagery Using Deep Learning\n\n- [e-Farmerce-platform](https://github.com/efarmerce/e-Farmerce-platform) -\u003e classify crop type\n\n- [sentinel2-deep-learning](https://github.com/d-smit/sentinel2-deep-learning) -\u003e Novel Training Methodologies for Land Classification of Sentinel-2 Imagery\n\n- [RSSC-transfer](https://github.com/risojevicv/RSSC-transfer) -\u003e The Role of Pre-Training in High-Resolution Remote Sensing Scene Classification\n\n- [Classifying Geo-Referenced Photos and Satellite Images for Supporting Terrain Classification](https://github.com/jorgemspereira/Classifying-Geo-Referenced-Photos) -\u003e detect floods\n\n- [Pay-More-Attention](https://github.com/williamzhao95/Pay-More-Attention) -\u003e Remote Sensing Image Scene Classification Based on an Enhanced Attention Module\n\n- [Remote Sensing Image Classification via Improved Cross-Entropy Loss and Transfer Learning Strategy Based on Deep Convolutional Neural Networks](https://github.com/AliBahri94/Remote-Sensing-Image-Classification-via-Improved-Cross-Entropy-Loss-and-Transfer-Learning-Strategy)\n\n- [DenseNet40-for-HRRSISC](https://github.com/BiQiWHU/DenseNet40-for-HRRSISC) -\u003e DenseNet40 for remote sensing image scene classification, uses UC Merced Dataset\n\n- [SKAL](https://github.com/hw2hwei/SKAL) -\u003e Looking Closer at the Scene: Multiscale Representation Learning for Remote Sensing Image Scene Classification\n\n- [potsdam-tensorflow-practice](https://github.com/medicinely/potsdam-tensorflow-practice) -\u003e image classification of Potsdam dataset using tensorflow\n\n- [SAFF](https://github.com/zh-hike/SAFF) -\u003e Self-Attention-Based Deep Feature Fusion for Remote Sensing Scene Classification\n\n- [GLNET](https://github.com/wuchangsheng951/GLNET) -\u003e Convolutional Neural Networks Based Remote Sensing Scene Classification under Clear and Cloudy Environments\n\n- [Remote-sensing-image-classification](https://github.com/hiteshK03/Remote-sensing-image-classification) -\u003e transfer learning using pytorch to classify remote sensing data into three classes: aircrafts, ships, none\n\n- [remote_sensing_pretrained_models](https://github.com/lsh1994/remote_sensing_pretrained_models) -\u003e as an alternative to fine tuning on models pretrained on ImageNet, here some CNN are pretrained on the RSD46-WHU \u0026 AID datasets\n\n- [CNN_AircraftDetection](https://github.com/UKMIITB/CNN_AircraftDetection) -\u003e CNN for aircraft detection in satellite images using keras\n\n- [OBIC-GCN](https://github.com/CVEO/OBIC-GCN) -\u003e Object-based Classification Framework of Remote Sensing Images with Graph Convolutional Networks\n\n- [aitlas-arena](https://github.com/biasvariancelabs/aitlas-arena) -\u003e An open-source benchmark framework for evaluating state-of-the-art deep learning approaches for image classification in Earth Observation (EO)\n\n- [droughtwatch](https://github.com/wandb/droughtwatch) -\u003e Satellite-based Prediction of Forage Conditions for Livestock in Northern Kenya\n\n- [JSTARS_2020_DPN-HRA](https://github.com/B-Xi/JSTARS_2020_DPN-HRA) -\u003e Deep Prototypical Networks With Hybrid Residual Attention for Hyperspectral Image Classification\n\n- [SIGNA](https://github.com/kyle-one/SIGNA) -\u003e Semantic Interleaving Global Channel Attention for Multilabel Remote Sensing Image Classification\n\n- [PBDL](https://github.com/Usman1021/PBDL) -\u003e Patch-Based Discriminative Learning for Remote Sensing Scene Classification\n\n- [EmergencyNet](https://github.com/ckyrkou/EmergencyNet) -\u003e identify fire and other emergencies from a drone\n\n- [satellite-deforestation](https://github.com/drewhibbard/satellite-deforestation) -\u003e Using Satellite Imagery to Identify the Leading Indicators of Deforestation, applied to the Kaggle Challenge Understanding the Amazon from Space\n\n- [RSMLC](https://github.com/marjanstoimchev/RSMLC) -\u003e Deep Network Architectures as Feature Extractors for Multi-Label Classification of Remote Sensing Images\n\n- [FireRisk](https://github.com/CharmonyShen/FireRisk) -\u003e A Remote Sensing Dataset for Fire Risk Assessment with Benchmarks Using Supervised and Self-supervised Learning\n\n- [flood_susceptibility_mapping](https://github.com/omarseleem92/flood_susceptibility_mapping) -\u003e Towards urban flood susceptibility mapping using data-driven models in Berlin, Germany\n\n- [tick-tick-bloom](https://github.com/drivendataorg/tick-tick-bloom) -\u003e Winners of the Tick Tick Bloom: Harmful Algal Bloom Detection Challenge. Task was to predict severity of algae bloom, winners used decision trees\n\n- [Estimating coal power plant operation from satellite images with computer vision](https://transitionzero.medium.com/estimating-coal-power-plant-operation-from-satellite-images-with-computer-vision-b966af56919e) -\u003e use Sentinel 2 data to identify if a coal power plant is on or off, with dataset and repo\n\n- [Building-detection-and-roof-type-recognition](https://github.com/loosgagnet/Building-detection-and-roof-type-recognition) -\u003e A CNN-Based Approach for Automatic Building Detection and Recognition of Roof Types Using a Single Aerial Image\n\n- [Performance Comparison of Multispectral Channels for Land Use Classification](https://github.com/tejasri19/EuroSAT_data_analysis) -\u003e Implemented ResNet-50, ResNet-101, ResNet-152, Vision Transformer on RGB and multispectral versions of EuroSAT dataset.\n\n- [SNN4Space](https://github.com/AndrzejKucik/SNN4Space) -\u003e project which investigates the feasibility of deploying spiking neural networks (SNN) in land cover and land use classification tasks\n\n- [vessel-classification](https://github.com/GlobalFishingWatch/vessel-classification) -\u003e classify vessels and identify fishing behavior based on AIS data\n\n- [RSMamba](https://github.com/KyanChen/RSMamba) -\u003e Remote Sensing Image Classification with State Space Model\n\n- [BirdSAT](https://github.com/mvrl/BirdSAT) -\u003e Cross-View Contrastive Masked Autoencoders for Bird Species Classification and Mapping\n\n- [EGNNA_WND](https://github.com/stevinc/EGNNA_WND) -\u003e Estimating the presence of the West Nile Disease employing Graph Neural network\n\n- [cyfi](https://github.com/drivendataorg/cyfi) -\u003e Estimate cyanobacteria density based on Sentinel-2 satellite imagery\n\n- [3DGAN-ViT](https://github.com/aj1365/3DGAN-ViT) -\u003e A deep learning framework based on generative adversarial networks and vision transformer for complex wetland classification puplished in [International Journal of Applied Earth Observation and Geoinformation](https://www.sciencedirect.com/science/article/pii/S1569843222002837)\n\n- [EfficientBigEarthNet](https://github.com/Orion-AI-Lab/EfficientBigEarthNet) -\u003e Code and models from the paper [Benchmarking and scaling of deep learning models for land cover image classification](https://www.sciencedirect.com/science/article/pii/S0924271622003057).\n\n- [automatic_solar_pv_detection](https://github.com/KennSmithDS/automatic_solar_pv_detection) -\u003e Automatic Solar PV Panel Image Classification with Deep Neural Network Transfer Learning\n\n- [U-netR](https://github.com/JonathanVSV/U-netR) -\u003e Land Use Land Cover Classification with U-Net: Advantages of Combining Sentinel-1 and Sentinel-2 Imagery [paper](https://doi.org/10.3390/rs13183600)\n\n#\n## Segmentation\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/segmentation.png\" width=\"500\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003e(left) a satellite image and (right) the semantic classes in the image.\u003c/b\u003e\n\u003c/p\u003e\n\nImage segmentation is a crucial step in image analysis and computer vision, with the goal of dividing an image into semantically meaningful segments or regions. The process of image segmentation assigns a class label to each pixel in an image, effectively transforming an image from a 2D grid of pixels into a 2D grid of pixels with assigned class labels. One common application of image segmentation is road or building segmentation, where the goal is to identify and separate roads and buildings from other features within an image. To accomplish this task, single class models are often trained to differentiate between roads and background, or buildings and background. These models are designed to recognize specific features, such as color, texture, and shape, that are characteristic of roads or buildings, and use this information to assign class labels to the pixels in an image. Another common application of image segmentation is land use or crop type classification, where the goal is to identify and map different land cover types within an image. In this case, multi-class models are typically used to recognize and differentiate between multiple classes within an image, such as forests, urban areas, and agricultural land. These models are capable of recognizing complex relationships between different land cover types, allowing for a more comprehensive understanding of the image content. Read [A brief introduction to satellite image segmentation with neural networks](https://medium.com/@robmarkcole/a-brief-introduction-to-satellite-image-segmentation-with-neural-networks-33ea732d5bce). **Note** that many articles which refer to 'hyperspectral land classification' are often actually describing semantic segmentation.\n\n### Segmentation - Land use \u0026 land cover\n\n- [nga-deep-learning](https://github.com/jordancaraballo/nga-deep-learning) -\u003e performs semantic segmentation on high resultion GeoTIF data using a modified U-Net \u0026 Keras, published by NASA researchers\n\n- [Automatic Detection of Landfill Using Deep Learning](https://github.com/AnupamaRajkumar/LandfillDetection_SemanticSegmentation)\n\n- [SpectralNET](https://github.com/tanmay-ty/SpectralNET) -\u003e a 2D wavelet CNN for Hyperspectral Image Classification, uses Salinas Scene dataset \u0026 Keras\n\n- [laika](https://github.com/datasciencecampus/laika) -\u003e The goal of this repo is to research potential sources of satellite image data and to implement various algorithms for satellite image segmentation\n\n- landcover dot io -\u003e was a human-in-the-loop AI tool to drastically reduce the time required to produce an accurate Land Use/Land Cover (LULC) map, [blog post](http://devseed.com/blog/2021-05-17-pearl-ai-land-cover), used Microsoft Planetary Computer and ML models run locally in the browser. Code for [backelnd](https://github.com/developmentseed/pearl-backend) and [frontend](https://github.com/developmentseed/pearl-frontend)\n\n- [Land Cover Classification with U-Net](https://baratam-tarunkumar.medium.com/land-cover-classification-with-u-net-aa618ea64a1b) -\u003e Satellite Image Multi-Class Semantic Segmentation Task with PyTorch Implementation of U-Net, uses DeepGlobe Land Cover Segmentation dataset, with [code](https://github.com/TarunKumar1995-glitch/land_cover_classification_unet)\n\n- [Multi-class semantic segmentation of satellite images using U-Net](https://github.com/rogerxujiang/dstl_unet) using DSTL dataset, tensorflow 1 \u0026 python 2.7. \n\n- [Codebase for multi class land cover classification with U-Net](https://github.com/jaeeolma/lulc_ml) accompanying a masters thesis, uses Keras\n\n- [dubai-satellite-imagery-segmentation](https://github.com/ayushdabra/dubai-satellite-imagery-segmentation) -\u003e due to the small dataset, image augmentation was used\n\n- [CDL-Segmentation](https://github.com/asimniazi63/CDL-Segmentation) -\u003e Deep Learning Based Land Cover and Crop Type Classification: A Comparative Study. Compares UNet, SegNet \u0026 DeepLabv3+\n\n- [LoveDA](https://github.com/Junjue-Wang/LoveDA) -\u003e A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation\n\n- [Satellite Imagery Semantic Segmentation with CNN](https://joshting.medium.com/satellite-imagery-segmentation-with-convolutional-neural-networks-f9254de3b907) -\u003e 7 different segmentation classes, DeepGlobe Land Cover Classification Challenge dataset, with [repo](https://github.com/justjoshtings/satellite_image_segmentation)\n\n- [Aerial Semantic Segmentation using U-Net Deep Learning Model](https://medium.com/@rehman.aimal/aerial-semantic-segmentation-using-u-net-deep-learning-model-3356a53c915f) medium article, with [repo](https://github.com/aimalrehman92/Multiclass-Semantic-Segmentation-with-U-NET)\n\n- [UNet-Satellite-Image-Segmentation](https://github.com/YudeWang/UNet-Satellite-Image-Segmentation) -\u003e A Tensorflow implentation of light UNet semantic segmentation framework\n\n- [DeepGlobe Land Cover Classification Challenge solution](https://github.com/GeneralLi95/deepglobe_land_cover_classification_with_deeplabv3plus)\n\n- [Semantic-segmentation-with-PyTorch-Satellite-Imagery](https://github.com/JenAlchimowicz/Semantic-segmentation-with-PyTorch-Satellite-Imagery) -\u003e predict 25 classes on RGB imagery taken to assess the damage after Hurricane Harvey\n\n- [Semantic Segmentation With Sentinel-2 Imagery](https://github.com/pavlo-seimskyi/semantic-segmentation-satellite-imagery) -\u003e uses LandCoverNet dataset and fast.ai\n\n- [CNN_Enhanced_GCN](https://github.com/qichaoliu/CNN_Enhanced_GCN) -\u003e CNN-Enhanced Graph Convolutional Network With Pixel- and Superpixel-Level Feature Fusion for Hyperspectral Image Classification\n\n- [LULCMapping-WV3images-CORINE-DLMethods](https://github.com/esertel/LULCMapping-WV3images-CORINE-DLMethods) -\u003e Land Use and Land Cover Mapping Using Deep Learning Based Segmentation Approaches and VHR Worldview-3 Images\n\n- [MCANet](https://github.com/yisun98/SOLC) -\u003e A joint semantic segmentation framework of optical and SAR images for land use classification. Uses [WHU-OPT-SAR-dataset](https://github.com/AmberHen/WHU-OPT-SAR-dataset)\n\n- [MUnet-LUC](https://github.com/abhi170599/MUnet-LUC)\n\n- [land-cover](https://github.com/lucashu1/land-cover) -\u003e Model Generalization in Deep Learning Applications for Land Cover Mapping\n\n- [generalizablersc](https://github.com/dgominski/generalizablersc) -\u003e Cross-dataset Learning for Generalizable Land Use Scene Classification\n\n- [Large-scale-Automatic-Identification-of-Urban-Vacant-Land](https://github.com/SkydustZ/Large-scale-Automatic-Identification-of-Urban-Vacant-Land) -\u003e Large-scale automatic identification of urban vacant land using semantic segmentation of high-resolution remote sensing images\n\n- [SSLTransformerRS](https://github.com/HSG-AIML/SSLTransformerRS) -\u003e Self-supervised Vision Transformers for Land-cover Segmentation and\n  Classification\n\n- [aerial-tile-segmentation](https://github.com/mrsebai/aerial-tile-segmentation) -\u003e Large satellite image semantic segmentation into 6 classes using Tensorflow 2.0 and ISPRS benchmark dataset\n\n- [LULCMapping-WV3images-CORINE-DLMethods](https://github.com/burakekim/LULCMapping-WV3images-CORINE-DLMethods) -\u003e Land Use and Land Cover Mapping Using Deep Learning Based Segmentation Approaches and VHR Worldview-3 Images\n\n- [DCSA-Net](https://github.com/Julia90/DCSA-Net) -\u003e Dynamic Convolution Self-Attention Network for Land-Cover Classification in VHR Remote-Sensing Images\n\n- [CHeGCN-CNN_enhanced_Heterogeneous_Graph](https://github.com/Liuzhizhiooo/CHeGCN-CNN_enhanced_Heterogeneous_Graph) -\u003e CNN-Enhanced Heterogeneous Graph Convolutional Network: Inferring Land Use from Land Cover with a Case Study of Park Segmentation\n\n- [TCSVT_2022_DGSSC](https://github.com/B-Xi/TCSVT_2022_DGSSC) -\u003e DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery\n\n- [DeepForest-Wetland-Paper](https://github.com/aj1365/DeepForest-Wetland-Paper) -\u003e Deep Forest classifier for wetland mapping using the combination of Sentinel-1 and Sentinel-2 data, GIScience \u0026 Remote Sensing\n\n- [Wetland_UNet](https://github.com/conservation-innovation-center/Wetland_UNet) -\u003e UNet models that can delineate wetlands using remote sensing data input including bands from Sentinel-2 LiDAR and geomorphons. By the Conservation Innovation Center of Chesapeake Conservancy and Defenders of Wildlife\n\n- [DPA](https://github.com/x-ytong/DPA) -\u003e DPA is an unsupervised domain adaptation (UDA) method applied to different satellite images for larg-scale land cover mapping.\n\n- [dynamicworld](https://github.com/google/dynamicworld) -\u003e Dynamic World, Near real-time global 10 m land use land cover mapping\n\n- [spada](https://github.com/links-ads/spada) -\u003e Land Cover Segmentation with Sparse Annotations from Sentinel-2 Imagery\n\n- [M3SPADA](https://github.com/ecapliez/M3SPADA) -\u003e  Multi-Sensor Temporal Unsupervised Domain Adaptation for Land Cover Mapping with spatial pseudo labelling and adversarial learning\n\n- [GLNet](https://github.com/VITA-Group/GLNet) -\u003e Collaborative Global-Local Networks for Memory-Efficient Segmentation of Ultra-High Resolution Images\n\n- [LoveNAS](https://github.com/Junjue-Wang/LoveNAS) -\u003e LoveNAS: Towards Multi-Scene Land-Cover Mapping via Hierarchical Searching Adaptive Network\n\n- [FLAIR-2 challenge](https://github.com/IGNF/FLAIR-2) -\u003e Semantic segmentation and domain adaptation challenge proposed by the French National Institute of Geographical and Forest Information (IGN)\n\n- [flair-2 8th place solution](https://github.com/association-rosia/flair-2)\n\n- [igarss-spada](https://github.com/links-ads/spada) -\u003e Dataset and code for the paper Land Cover Segmentation with Sparse Annotations from Sentinel-2 Imagery [IGARSS 2023](https://arxiv.org/abs/2306.16252). \n\n### Segmentation - Vegetation, deforestation, crops \u0026 crop boundaries\n\nNote that deforestation detection may be treated as a segmentation task or a change detection task\n\n- [DetecTree](https://github.com/martibosch/detectree) -\u003e Tree detection from aerial imagery in Python, a LightGBM classifier of tree/non-tree pixels from aerial imagery\n\n- [Сrор field boundary detection: approaches and main challenges](https://medium.com/geekculture/%D1%81r%D0%BE%D1%80-field-boundary-detection-approaches-and-main-challenges-46e37dd276bc) -\u003e Medium article, covering historical and modern approaches\n\n- [kenya-crop-mask](https://github.com/nasaharvest/kenya-crop-mask) -\u003e Annual and in-season crop mapping in Kenya - LSTM classifier to classify pixels as containing crop or not, and a multi-spectral forecaster that provides a 12 month time series given a partial input. Dataset downloaded from GEE and pytorch lightning used for training\n\n- [Tree species classification from from airborne LiDAR and hyperspectral data using 3D convolutional neural networks](https://github.com/jaeeolma/tree-detection-evo)\n\n- [crop-type-classification](https://medium.com/nerd-for-tech/crop-type-classification-cf5cc2593396) -\u003e using Sentinel 1 \u0026 2 data with a U-Net + LSTM, more features (i.e. bands) and higher resolution produced better results (article, no code)\n\n- [Find sports fields using Mask R-CNN and overlay on open-street-map](https://github.com/jremillard/images-to-osm)\n\n- [An LSTM to generate a crop mask for Togo](https://github.com/nasaharvest/togo-crop-mask)\n\n- [DeepSatModels](https://github.com/michaeltrs/DeepSatModels) -\u003e Context-self contrastive pretraining for crop type semantic segmentation\n\n- [farm-pin-crop-detection-challenge](https://github.com/simongrest/farm-pin-crop-detection-challenge) -\u003e Using eo-learn and fastai to identify crops from multi-spectral remote sensing data\n\n- [Detecting Agricultural Croplands from Sentinel-2 Satellite Imagery](https://medium.com/radiant-earth-insights/detecting-agricultural-croplands-from-sentinel-2-satellite-imagery-a025735d3bd8) -\u003e We developed UNet-Agri, a benchmark machine learning model that classifies croplands using open-access Sentinel-2 imagery at 10m spatial resolution\n\n- [DeepTreeAttention](https://github.com/weecology/DeepTreeAttention) -\u003e Implementation of Hang et al. 2020 \"Hyperspectral Image Classification with Attention Aided CNNs\" for tree species prediction\n\n- [Crop-Classification](https://github.com/bhavesh907/Crop-Classification) -\u003e crop classification using multi temporal satellite images\n\n- [ParcelDelineation](https://github.com/sustainlab-group/ParcelDelineation) -\u003e using a French polygons dataset and unet in keras\n\n- [crop-mask](https://github.com/nasaharvest/crop-mask) -\u003e End-to-end workflow for generating high resolution cropland maps, uses GEE \u0026 LSTM model\n\n- [DeepCropMapping](https://github.com/Lab-IDEAS/DeepCropMapping) -\u003e A multi-temporal deep learning approach with improved spatial generalizability for dynamic corn and soybean mapping, uses LSTM\n\n- [ResUnet-a](https://github.com/Akhilesh64/ResUnet-a) -\u003e a deep learning framework for semantic segmentation of remotely sensed data\n\n- [DSD_paper_2020](https://github.com/JacobJeppesen/DSD_paper_2020) -\u003e Crop Type Classification based on Machine Learning with Multitemporal Sentinel-1 Data\n\n- [MR-DNN](https://github.com/yasir2afaq/Multi-resolution-deep-neural-network) -\u003e extract rice field from Landsat 8 satellite imagery\n\n- [deep_learning_forest_monitoring](https://github.com/waldeland/deep_learning_forest_monitoring) -\u003e Forest mapping and monitoring of the African continent using Sentinel-2 data and deep learning\n\n- [global-cropland-mapping](https://github.com/Charly-tian/global-cropland-mapping) -\u003e global multi-temporal cropland mapping\n\n- [U-Net for Semantic Segmentation of Soyabean Crop Fields with SAR images](https://joaootavionf007.medium.com/u-net-for-semantic-segmentation-of-soyabeans-crop-fields-with-sar-images-604232e49315)\n\n- [UNet-RemoteSensing](https://github.com/aryanVijaywargia/UNet-RemoteSensing) -\u003e uses 7 bands of Landsat and keras\n\n- [Landuse_DL](https://github.com/yghlc/Landuse_DL) -\u003e delineate landforms due to the thawing of ice-rich permafrost\n\n- [canopy](https://github.com/jonathanventura/canopy) -\u003e A Convolutional Neural Network Classifier Identifies Tree Species in Mixed-Conifer Forest from Hyperspectral Imagery\n\n- [RandomForest-Classification](https://github.com/florianbeyer/RandomForest-Classification) -\u003e Multisensor data to derive peatland vegetation communities using a fixed-wing unmanned aerial vehicle\n\n- [forest_change_detection](https://github.com/QuantuMobileSoftware/forest_change_detection) -\u003e forest change segmentation with time-dependent models, including Siamese, UNet-LSTM, UNet-diff, UNet3D models\n\n- [cultionet](https://github.com/jgrss/cultionet) -\u003e segmentation of cultivated land, built on PyTorch Geometric and PyTorch Lightning\n\n- [sentinel-tree-cover](https://github.com/wri/sentinel-tree-cover) -\u003e A global method to identify trees outside of closed-canopy forests with medium-resolution satellite imagery\n\n- [crop-type-detection-ICLR-2020](https://github.com/RadiantMLHub/crop-type-detection-ICLR-2020) -\u003e Winning Solutions from Crop Type Detection Competition at CV4A workshop, ICLR 2020\n\n- [Crop identification using satellite imagery](https://write.agrevolution.in/crop-identification-using-satellite-imagery-introduction-83d79344f9ee) -\u003e Medium article, introduction to crop identification\n\n- [S4A-Models](https://github.com/Orion-AI-Lab/S4A-Models) -\u003e Various experiments on the Sen4AgriNet dataset\n\n- [attention-mechanism-unet](https://github.com/davej23/attention-mechanism-unet) -\u003e An attention-based U-Net for detecting deforestation within satellite sensor imagery\n\n- [Cocoa_plantations_detection](https://github.com/antoine-spahr/Cocoa_plantations_detection) -\u003e Detecting cocoa plantation in Ivory Coast using Sentinel-2 remote sensing data using KNN, SVM, Random Forest and MLP\n\n- [SummerCrop_Deeplearning](https://github.com/AgriRS/SummerCrop_Deeplearning) -\u003e A Transferable Learning Classification Model and Carbon Sequestration Estimation of Crops in Farmland Ecosystem\n\n- [DeepForest](https://deepforest.readthedocs.io/en/latest/index.html) is a python package for training and predicting individual tree crowns from airborne RGB imagery\n\n- [Official repository for the \"Identifying trees on satellite images\" challenge from Omdena](https://github.com/cienciaydatos/ai-challenge-trees)\n\n- [Counting-Trees-using-Satellite-Images](https://github.com/A2Amir/Counting-Trees-using-Satellite-Images) -\u003e create an inventory of incoming and outgoing trees for an annual tree inspections, uses keras \u0026 semantic segmentation\n\n- [2020 Nature paper - An unexpectedly large count of trees in the West African Sahara and Sahel](https://www.nature.com/articles/s41586-020-2824-5) -\u003e tree detection framework based on U-Net \u0026 tensorflow 2 with code [here](https://github.com/ankitkariryaa/An-unexpectedly-large-count-of-trees-in-the-western-Sahara-and-Sahel/tree/v1.0.0)\n\n- [TreeDetection](https://github.com/AmirNiaraki/TreeDetection) -\u003e A color-based classifier to detect the trees in google image data along with tree visual localization and crown size calculations via OpenCV\n\n- [PTDM](https://github.com/hr8yhtzb/PTDM) -\u003e Pomelo Tree Detection Method Based on Attention Mechanism and Cross-Layer Feature Fusion\n\n- [urban-tree-detection](https://github.com/jonathanventura/urban-tree-detection) -\u003e Individual Tree Detection in Large-Scale Urban Environments using High-Resolution Multispectral Imagery. With [dataset](https://github.com/jonathanventura/urban-tree-detection-data)\n\n- [BioMassters_baseline](https://github.com/fnands/BioMassters_baseline) -\u003e a basic pytorch lightning baseline using a UNet for getting started with the [BioMassters challenge](https://www.drivendata.org/competitions/99/biomass-estimation/) (biomass estimation)\n\n- [Biomassters winners](https://github.com/drivendataorg/the-biomassters) -\u003e top 3 solutions\n\n- [kbrodt biomassters solution](https://github.com/kbrodt/biomassters) -\u003e 1st place solution\n\n- [quqixun biomassters solution](https://github.com/quqixun/BioMassters)\n\n- [biomass-estimation](https://github.com/azavea/biomass-estimation) -\u003e from Azavea, applied to Sentinel 1 \u0026 2\n\n- [3DUNetGSFormer](https://github.com/aj1365/3DUNetGSFormer) -\u003e A deep learning pipeline for complex wetland mapping using generative adversarial networks and Swin transformer\n\n- [SEANet_torch](https://github.com/long123524/SEANet_torch) -\u003e Using a semantic edge-aware multi-task neural network to delineate agricultural parcels from remote sensing images\n\n- [arborizer](https://github.com/RaffiBienz/arborizer) -\u003e Tree crowns segmentation and classification\n\n- [ReUse](https://github.com/priamus-lab/ReUse) -\u003e REgressive Unet for Carbon Storage and Above-Ground Biomass Estimation\n\n- [unet-sentinel](https://github.com/eliasqueirogavieira/unet-sentinel) -\u003e UNet to handle Sentinel-1 SAR images to identify deforestation\n\n- [MaskedSST](https://github.com/HSG-AIML/MaskedSST) -\u003e Masked Vision Transformers for Hyperspectral Image Classification\n\n- [UNet-defmapping](https://github.com/bragagnololu/UNet-defmapping) -\u003e master's thesis using UNet to map deforestation using Sentinel-2 Level 2A images, applied to Amazon and Atlantic Rainforest dataset\n\n- [cvpr-multiearth-deforestation-segmentation](https://github.com/h2oai/cvpr-multiearth-deforestation-segmentation) -\u003e multimodal Unet entry to the CVPR Multiearth 2023 deforestation challenge\n\n- [supervised-wheat-classification-using-pytorchs-torchgeo](https://medium.com/@sulemanhamdani10/supervised-wheat-classification-using-pytorchs-torchgeo-combining-satellite-imagery-and-python-fc7f95c82e) -\u003e supervised wheat classification using torchgeo\n\n- [TransUNetplus2](https://github.com/aj1365/TransUNetplus2) -\u003e TransU-Net++: Rethinking attention gated TransU-Net for deforestation mapping. Uses the Amazon and Atlantic forest dataset\n\n- [A high-resolution canopy height model of the Earth](https://github.com/langnico/global-canopy-height-model#a-high-resolution-canopy-height-model-of-the-earth) -\u003e A high-resolution canopy height model of the Earth\n\n- [Radiant Earth Spot the Crop Challenge](https://github.com/radiantearth/spot-the-crop-challenge) -\u003e Winning models from the Radiant Earth Spot the Crop Challenge, uses a time-series of Sentinel-2 multispectral data to classify crops in the Western Cape of South Africa. [Another solution](https://github.com/DariusTheGeek/Radiant-Earth-Spot-the-Crop-XL-Challenge)\n\n- [transfer-field-delineation](https://github.com/kerner-lab/transfer-field-delineation) -\u003e Multi-Region Transfer Learning for Segmentation of Crop Field Boundaries in Satellite Images with Limited Labels\n\n- [mowing-detection](https://github.com/lucas-batier/mowing-detection) -\u003e Automatic detection of mowing and grazing from Sentinel images\n\n- [PTAViT3D and PTAViT3DCA](https://github.com/feevos/tfcl) -\u003e Tackling fluffy clouds: field boundaries detection using time series of S2 and/or S1 imagery\n\n- [ai4boundaries](https://github.com/waldnerf/ai4boundaries) -\u003e a Python package that facilitates download of the AI4boundaries data set\n\n- [Nasa_harvest_field_boundary_competition](https://github.com/radiantearth/Nasa_harvest_field_boundary_competition) -\u003e Nasa Harvest Rwanda Field Boundary Detection Challenge Tutorial\n\n- [UTB_codes](https://github.com/zhu-xlab/UTB_codes) -\u003e The Urban Tree Canopy Cover in Brazil [article](https://nkszjx.github.io/projects/UTB.html)\n\n- [nasa_harvest_boundary_detection_challenge](https://github.com/geoaigroup/nasa_harvest_boundary_detection_challenge) -\u003e the 4th place solution for NASA Harvest Field Boundary Detection Challenge on Zindi.\n\n- [rainforest-segmentation](https://github.com/jcblsn/rainforest-segmentation) -\u003e Identifying and tracking deforestation in the Amazon Rainforest using state-of-the-art deep learning models and multispectral satellite imagery.\n\n- [Delineate Anything: Resolution-Agnostic Field Boundary Delineation on Satellite Imagery](https://github.com/Lavreniuk/Delineate-Anything)\n\n- [Semantic_segmentation_for_LCLUC](https://github.com/waterdmd/Semantic_segmentation_for_LCLUC) -\u003e Semantic Segmentation for Simultaneous Crop and Land Cover Land Use Classification Using Multi-Temporal Landsat Imagery\n\n### Segmentation - Water, coastlines, rivers \u0026 floods\n\n- [pytorch-waterbody-segmentation](https://github.com/gauthamk02/pytorch-waterbody-segmentation) -\u003e UNET model trained on the Satellite Images of Water Bodies dataset from Kaggle. The model is deployed on Hugging Face Spaces\n\n- [Flood Detection and Analysis using UNET with Resnet-34 as the back bone](https://github.com/orion29/Satellite-Image-Segmentation-for-Flood-Damage-Analysis) uses fastai\n\n- [Automatic Flood Detection from Satellite Images Using Deep Learning](https://medium.com/@omercaliskan99/automatic-flood-detection-from-satellite-images-using-deep-learning-f14fafd369e0)\n\n- [Semi-Supervised Classification and Segmentation on High Resolution Aerial Images - Solving the FloodNet problem](https://sahilkhose.medium.com/paper-presentation-e9bd0f3fb0bf)\n\n- [Houston_flooding](https://github.com/Lichtphyz/Houston_flooding) -\u003e labeling each pixel as either flooded or not using data from Hurricane Harvey. Dataset consisted of pre and post flood images, and a ground truth floodwater mask was created using unsupervised clustering (with DBScan) of image pixels with human cluster verification/adjustment\n\n- [ml4floods](https://github.com/spaceml-org/ml4floods) -\u003e An ecosystem of data, models and code pipelines to tackle flooding with ML\n\n- [A comprehensive guide to getting started with the ETCI Flood Detection competition](https://medium.com/cloud-to-street/jumpstart-your-machine-learning-satellite-competition-submission-2443b40d0a5a) -\u003e using Sentinel1 SAR \u0026 pytorch\n\n- [Map Floodwater of SAR Imagery with SageMaker](https://github.com/JayThibs/map-floodwater-sar-imagery-on-sagemaker) -\u003e applied to Sentinel-1 dataset\n\n- [1st place solution for STAC Overflow: Map Floodwater from Radar Imagery hosted by Microsoft AI for Earth](https://github.com/sweetlhare/STAC-Overflow) -\u003e combines Unet with Catboostclassifier, taking their maxima, not the average\n\n- [hydra-floods](https://github.com/Servir-Mekong/hydra-floods) -\u003e an open source Python application for downloading, processing, and delivering surface water maps derived from remote sensing data\n\n- [CoastSat](https://github.com/kvos/CoastSat) -\u003e tool for mapping coastlines which has an extension [CoastSeg](https://github.com/dbuscombe-usgs/CoastSeg) using segmentation models\n\n- [Satellite_Flood_Segmentation_of_Harvey](https://github.com/morgan-tam/Satellite_Flood_Segmentation_of_Harvey) -\u003e explores both deep learning and traditional kmeans\n\n- [Flood Event Detection Utilizing Satellite Images](https://github.com/KonstantinosF/Flood-Detection---Satellite-Images)\n\n- [ETCI-2021-Competition-on-Flood-Detection](https://github.com/sidgan/ETCI-2021-Competition-on-Flood-Detection) -\u003e Experiments on Flood Segmentation on Sentinel-1 SAR Imagery with Cyclical Pseudo Labeling and Noisy Student Training\n\n- [FDSI](https://github.com/keillernogueira/FDSI) -\u003e Flood Detection in Satellite Images - 2017 Multimedia Satellite Task\n\n- [deepwatermap](https://github.com/isikdogan/deepwatermap) -\u003e a deep model that segments water on multispectral images\n\n- [rivamap](https://github.com/isikdogan/rivamap) -\u003e an automated river analysis and mapping engine\n\n- [deep-water](https://github.com/maxbeber/deep-water) -\u003e track changes in water level\n\n- [WatNet](https://github.com/xinluo2018/WatNet) -\u003e A deep ConvNet for surface water mapping based on Sentinel-2 image, uses the [Earth Surface Water Dataset](https://zenodo.org/record/5205674#.YoMjyZPMK3I)\n\n- [A-U-Net-for-Flood-Extent-Mapping](https://github.com/jorgemspereira/A-U-Net-for-Flood-Extent-Mapping)\n\n- [floatingobjects](https://github.com/ESA-PhiLab/floatingobjects) -\u003e TOWARDS DETECTING FLOATING OBJECTS ON A GLOBAL SCALE WITHLEARNED SPATIAL FEATURES USING SENTINEL 2. Uses U-Net \u0026 pytorch\n\n- [SpaceNet8](https://github.com/SpaceNetChallenge/SpaceNet8) -\u003e baseline Unet solution to detect flooded roads and buildings\n\n- [dlsim](https://github.com/nyokoya/dlsim) -\u003e Breaking the Limits of Remote Sensing by Simulation and Deep Learning for Flood and Debris Flow Mapping\n\n- [Water-HRNet](https://github.com/faye0078/Water-Extraction) -\u003e HRNet trained on Sentinel 2\n\n- [semantic segmentation model to identify newly developed or flooded land](https://github.com/Azure/pixel_level_land_classification) using NAIP imagery provided by the Chesapeake Conservancy, training on MS Azure\n\n- [BandNet](https://github.com/IamShubhamGupto/BandNet) -\u003e Analysis and application of multispectral data for water segmentation using machine learning. Uses Sentinel-2 data\n\n- [mmflood](https://github.com/edornd/mmflood) -\u003e MMFlood: A Multimodal Dataset for Flood Delineation From Satellite Imagery (Sentinel 1 SAR)\n\n- [Urban_flooding](https://github.com/omarseleem92/Urban_flooding) -\u003e Towards transferable data-driven models to predict urban pluvial flood water depth in Berlin, Germany\n\n- [Flood-Mapping-Using-Satellite-Images](https://github.com/KonstantinosF/Flood-Mapping-Using-Satellite-Images) -\u003e masters thesis comparing Random Forest \u0026 Unet\n\n- [MECNet](https://github.com/zhilyzhang/MECNet) -\u003e Rich CNN features for water-body segmentation from very high resolution aerial and satellite imagery\n\n- [SWRNET](https://github.com/trongan93/swrnet) -\u003e A Deep Learning Approach for Small Surface Water Area Recognition Onboard Satellite\n\n- [elwha-segmentation](https://github.com/StefanTodoran/elwha-segmentation) -\u003e fine-tuning Meta's Segment Anything (SAM) for bird's eye view river pixel segmentation\n\n- [RiverSnap](https://github.com/ArminMoghimi/RiverSnap) -\u003e code for paper: A Comparative Performance Analysis of Popular Deep Learning Models and Segment Anything Model (SAM) for River Water Segmentation in Close-Range Remote Sensing Imagery\n\n- [SAR-water-segmentation](https://github.com/myeungun/SAR-water-segmentation) -\u003e Deep Learning based Water Segmentation Using KOMPSAT-5 SAR Images\n\n### Segmentation - Fire, smoke \u0026 burn areas\n\n- [SatelliteVu-AWS-Disaster-Response-Hackathon](https://github.com/SatelliteVu/SatelliteVu-AWS-Disaster-Response-Hackathon) -\u003e fire spread prediction using classical ML \u0026 deep learning\n\n- [Wild Fire Detection](https://github.com/yueureka/WildFireDetection) using U-Net trained on Databricks \u0026 Keras, semantic segmentation\n\n- [A Practical Method for High-Resolution Burned Area Monitoring Using Sentinel-2 and VIIRS](https://github.com/mnpinto/FireHR)\n\n- [IndustrialSmokePlumeDetection](https://github.com/HSG-AIML/IndustrialSmokePlumeDetection) -\u003e using Sentinel-2 \u0026 a modified ResNet-50\n\n- [burned-area-detection](https://github.com/dymaxionlabs/burned-area-detection) -\u003e uses Sentinel-2\n\n- [rescue](https://github.com/dbdmg/rescue) -\u003e Attention to fires: multi-channel deep-learning models forwildfire severity prediction\n\n- [smoke_segmentation](https://github.com/jeffwen/smoke_segmentation) -\u003e Segmenting smoke plumes and predicting density from GOES imagery\n\n- [wildfire-detection](https://github.com/amanbasu/wildfire-detection) -\u003e Using Vision Transformers for enhanced wildfire detection in satellite images\n\n- [Burned_Area_Detection](https://github.com/prhuppertz/Burned_Area_Detection) -\u003e Detecting Burned Areas with Sentinel-2 data\n\n- [burned-area-baseline](https://github.com/lccol/burned-area-baseline) -\u003e baseline unet model accompanying the Satellite Burned Area Dataset (Sentinel 1 \u0026 2)\n\n- [burned-area-seg](https://github.com/links-ads/burned-area-seg) -\u003e Burned area segmentation from Sentinel-2 using multi-task learning\n\n- [chabud2023](https://github.com/developmentseed/chabud2023) -\u003e Change detection for Burned area Delineation (ChaBuD) ECML/PKDD 2023 challenge\n\n- [Post Wildfire Burnt-up Detection using Siamese-UNet](https://github.com/kavyagupta/chabud) -\u003e on Chadbud dataset\n\n- [vit-burned-detection](https://github.com/DarthReca/vit-burned-detection) -\u003e Vision transformers in burned area delineation\n\n### Segmentation - Landslides\n\n- [landslide-sar-unet](https://github.com/iprapas/landslide-sar-unet) -\u003e Deep Learning for Rapid Landslide Detection using Synthetic Aperture Radar (SAR) Datacubes\n\n- [landslide-mapping-with-cnn](https://github.com/nprksh/landslide-mapping-with-cnn) -\u003e A new strategy to map landslides with a generalized convolutional neural network\n\n- [Relict_landslides_CNN_kmeans](https://github.com/SPAMLab/data_sharing/tree/main/Relict_landslides_CNN_kmeans) -\u003e Relict landslide detection in rainforest areas using a combination of k-means clustering algorithm and Deep-Learning semantic segmentation models\n\n- [Landslide-mapping-on-SAR-data-by-Attention-U-Net](https://github.com/lorenzonava96/Landslide-mapping-on-SAR-data-by-Attention-U-Net) -\u003e Rapid Mapping of landslide on SAR data by Attention U-net\n\n- [SAR-landslide-detection-pretraining](https://github.com/VMBoehm/SAR-landslide-detection-pretraining) -\u003e SAR-based landslide classification pretraining leads to better segmentation\n\n- [Landslide mapping from Sentinel-2 imagery through change detection](https://github.com/links-ads/igarss-landslide-delineation)\n\n- [landslide4sense-solution](https://github.com/iamtekson/landslide4sense-solution) -\u003e solution of Tek Kshetri\n\n### Segmentation - Glaciers\n\n- [HED-UNet](https://github.com/khdlr/HED-UNet) -\u003e a model for simultaneous semantic segmentation and edge detection, examples provided are glacier fronts and building footprints using the Inria Aerial Image Labeling dataset\n\n- [glacier_mapping](https://github.com/krisrs1128/glacier_mapping) -\u003e Mapping glaciers in the Hindu Kush Himalaya, Landsat 7 images, Shapefile labels of the glaciers, Unet with dropout\n\n- [glacier-detect-ML](https://github.com/mikeskaug/glacier-detect-ML) -\u003e a simple logistic regression model to identify a glacier in Landsat satellite imagery\n\n- [GlacierSemanticSegmentation](https://github.com/n9Mtq4/GlacierSemanticSegmentation)\n\n- [Antarctic-fracture-detection](https://github.com/chingyaolai/Antarctic-fracture-detection) -\u003e uses UNet with the MODIS Mosaic of Antarctica to detect surface fractures\n\n- [sentinel_lakeice](https://github.com/prs-eth/sentinel_lakeice) -\u003e Lake Ice Detection from Sentinel-1 SAR with Deep Learning\n\n### Segmentation - Other environmental\n\n- [Detection of Open Landfills](https://github.com/dymaxionlabs/basurales) -\u003e uses Sentinel-2 to detect large changes in the Normalized Burn Ratio (NBR)\n\n- [sea_ice_remote_sensing](https://github.com/sum1lim/sea_ice_remote_sensing) -\u003e Sea Ice Concentration classification\n\n- [Methane-detection-from-hyperspectral-imagery](https://github.com/satish1901/Methane-detection-from-hyperspectral-imagery) -\u003e Deep Remote Sensing Methods for Methane Detection in Overhead Hyperspectral Imagery\n\n- [methane-emission-project](https://github.com/stlbnmaria/methane-emission-project) -\u003e Classification CNNs was combined in an ensemble approach with traditional methods on tabular data\n\n- [CH4Net](https://github.com/annavaughan/CH4Net) -\u003e A fast, simple model for detection of methane plumes using sentinel-2\n\n- [EddyNet](https://github.com/redouanelg/EddyNet) -\u003e A Deep Neural Network For Pixel-Wise Classification of Oceanic Eddies\n\n- [schisto-vegetation](https://github.com/deleo-lab/schisto-vegetation) -\u003e Deep Learning Segmentation of Satellite Imagery Identifies Aquatic Vegetation Associated with Snail Intermediate Hosts of Schistosomiasis in Senegal, Africa\n\n- [Earthformer](https://github.com/amazon-science/earth-forecasting-transformer) -\u003e Exploring space-time transformers for earth system forecasting\n\n- [weather4cast-2022](https://github.com/iarai/weather4cast-2022) -\u003e Unet-3D baseline model for Weather4cast Rain Movie Prediction competition\n\n- [WeatherFusionNet](https://github.com/Datalab-FIT-CTU/weather4cast-2022) -\u003e Predicting Precipitation from Satellite Data. weather4cast-2022 1st place solution\n\n- [marinedebrisdetector](https://github.com/MarcCoru/marinedebrisdetector) -\u003e Large-scale Detection of Marine Debris in Coastal Areas with Sentinel-2\n\n- [kaggle-identify-contrails-4th](https://github.com/selimsef/kaggle-identify-contrails-4th) -\u003e 4th place Solution, Google Research - Identify Contrails to Reduce Global Warming\n\n- [MineSegSAT](https://github.com/macdonaldezra/MineSegSAT) -\u003e An automated system to evaluate mining disturbed area extents from Sentinel-2 imagery\n\n- [STARCOP: Semantic Segmentation of Methane Plumes with Hyperspectral Machine Learning models](https://github.com/spaceml-org/STARCOP)\n\n- [asos](https://gitlab.jsc.fz-juelich.de/kiste/asos) -\u003e Recognizing protected and anthropogenic patterns in landscapes using interpretable machine learning and satellite imagery\n\n### Segmentation - Roads \u0026 sidewalks\nExtracting roads is challenging due to the occlusions caused by other objects and the complex traffic environment\n\n- [ChesapeakeRSC](https://github.com/isaaccorley/ChesapeakeRSC) -\u003e segmentation to extract roads from the background but are additionally evaluated by how they perform on the \"Tree Canopy Over Road\" class\n\n- [ML_EPFL_Project_2](https://github.com/LucasBrazCappelo/ML_EPFL_Project_2) -\u003e U-Net in Pytorch to perform semantic segmentation of roads on satellite images\n\n- [Semantic Segmentation of roads](https://vihan-tyagi.medium.com/semantic-segmentation-of-satellite-images-based-on-deep-learning-algorithms-ea5ec408ac53) using  U-net Keras, OSM data, project summary article by student, no code\n\n- [Winning Solutions from SpaceNet Road Detection and Routing Challenge](https://github.com/SpaceNetChallenge/RoadDetector)\n\n- [RoadVecNet](https://github.com/gismodelling/RoadVecNet) -\u003e Road-Network-Segmentation-and-Vectorization in keras with dataset\n\n- [Detecting road and road types jupyter notebook](https://github.com/taspinar/sidl/blob/master/notebooks/2_Detecting_road_and_roadtypes_in_sattelite_images.ipynb)\n\n- [awesome-deep-map](https://github.com/antran89/awesome-deep-map) -\u003e A curated list of resources dedicated to deep learning / computer vision algorithms for mapping. The mapping problems include road network inference, building footprint extraction, etc.\n\n- [RoadTracer: Automatic Extraction of Road Networks from Aerial Images](https://github.com/mitroadmaps/roadtracer) -\u003e uses an iterative search process guided by a CNN-based decision function to derive the road network graph directly from the output of the CNN\n\n- [road_detection_mtl](https://github.com/ntelo007/road_detection_mtl) -\u003e Road Detection using a multi-task Learning technique to improve the performance of the road detection task by incorporating prior knowledge constraints, uses the SpaceNet Roads Dataset\n\n- [road_connectivity](https://github.com/anilbatra2185/road_connectivity) -\u003e Improved Road Connectivity by Joint Learning of Orientation and Segmentation (CVPR2019)\n\n- [Road-Network-Extraction using classical Image processing](https://github.com/abhaykes1/Road-Network-Extraction) -\u003e blur \u0026 canny edge detection\n\n- [SPIN_RoadMapper](https://github.com/wgcban/SPIN_RoadMapper) -\u003e Extracting Roads from Aerial Images via Spatial and Interaction Space Graph Reasoning for Autonomous Driving\n\n- [road_extraction_remote_sensing](https://github.com/jiankang1991/road_extraction_remote_sensing) -\u003e pytorch implementation, CVPR2018 DeepGlobe Road Extraction Challenge submission. See also [DeepGlobe-Road-Extraction-Challenge](https://github.com/zlckanata/DeepGlobe-Road-Extraction-Challenge)\n\n- [RoadDetections dataset by Microsoft](https://github.com/microsoft/RoadDetections)\n\n- [CoANet](https://github.com/mj129/CoANet) -\u003e Connectivity Attention Network for Road Extraction From Satellite Imagery. The CoA module incorporates graphical information to ensure the connectivity of roads are better preserved\n\n- [Satellite Imagery Road Segmentation](https://medium.com/@nithishmailme/satellite-imagery-road-segmentation-ad2964dc3812) -\u003e intro articule on Medium using the kaggle [Massachusetts Roads Dataset](https://www.kaggle.com/datasets/balraj98/massachusetts-roads-dataset)\n\n- [Label-Pixels](https://github.com/venkanna37/Label-Pixels) -\u003e for semantic segmentation of roads and other features\n\n- [Satellite-image-road-extraction](https://github.com/amanhari-projects/Satellite-image-road-extraction) -\u003e Road Extraction by Deep Residual U-Net\n\n- [road_building_extraction](https://github.com/jeffwen/road_building_extraction) -\u003e Pytorch implementation of U-Net architecture for road and building extraction\n\n- [RCFSNet](https://github.com/CVer-Yang/RCFSNet) -\u003e Road Extraction From Satellite Imagery by Road Context and Full-Stage Feature\n\n- [SGCN](https://github.com/tist0bsc/SGCN) -\u003e Split Depth-Wise Separable Graph-Convolution Network for Road Extraction in Complex Environments From High-Resolution Remote-Sensing Images\n\n- [ASPN](https://github.com/pshams55/ASPN) -\u003e Road Segmentation for Remote Sensing Images using Adversarial Spatial Pyramid Networks\n\n - [FCNs-for-road-extraction-keras](https://github.com/zetrun-liu/FCNs-for-road-extraction-keras) -\u003e Road extraction of high-resolution remote sensing images based on various semantic segmentation networks\n\n- [cresi](https://github.com/avanetten/cresi) -\u003e Road network extraction from satellite imagery, with speed and travel time estimates\n\n- [D-LinkNet](https://github.com/NekoApocalypse/road-extraction-d-linknet) -\u003e LinkNet with Pretrained Encoder and Dilated Convolution for High Resolution Satellite Imagery Road Extraction\n\n- [Sat2Graph](https://github.com/songtaohe/Sat2Graph) -\u003e Road Graph Extraction through Graph-Tensor Encoding\n\n- [Image-Segmentation)](https://github.com/mschulz/Image-Segmentation) -\u003e using Massachusetts Road dataset and fast.ai\n\n- [RoadTracer-M](https://github.com/astro-ck/RoadTracer-M) -\u003e Road Network Extraction from Satellite Images Using CNN Based Segmentation and Tracing\n\n- [ScRoadExtractor](https://github.com/weiyao1996/ScRoadExtractor) -\u003e Scribble-based Weakly Supervised Deep Learning for Road Surface Extraction from Remote Sensing Images\n\n- [RoadDA](https://github.com/LANMNG/RoadDA) -\u003e Stagewise Unsupervised Domain Adaptation with Adversarial Self-Training for Road Segmentation of Remote Sensing Images\n\n- [DeepSegmentor](https://github.com/yhlleo/DeepSegmentor) -\u003e A Pytorch implementation of DeepCrack and RoadNet projects\n\n- [Cascaded Residual Attention Enhanced Road Extraction from Remote Sensing Images](https://github.com/liaochengcsu/Cascade_Residual_Attention_Enhanced_for_Refinement_Road_Extraction)\n\n- [NL-LinkNet](https://github.com/SIAnalytics/nia-road-baseline) -\u003e Toward Lighter but More Accurate Road Extraction with Non-Local Operations\n\n- [IRSR-net](https://github.com/yangzhen1252/IRSR-net) -\u003e Lightweight Remote Sensing Road Detection Network\n\n- [hironex](https://github.com/johannesuhl/hironex) -\u003e A python tool for automatic, fully unsupervised extraction of historical road networks from historical maps\n\n- [Road_detection_model](https://github.com/JonasImazon/Road_detection_model) -\u003e Mapping Roads in the Brazilian Amazon with Artificial Intelligence and Sentinel-2\n\n- [DTnet](https://github.com/huzican695/DTnet) -\u003e Road detection via a dual-task network based on cross-layer graph fusion modules\n\n- [Automatic-Road-Extraction-from-Historical-Maps-using-Deep-Learning-Techniques](https://github.com/UrbanOccupationsOETR/Automatic-Road-Extraction-from-Historical-Maps-using-Deep-Learning-Techniques) -\u003e Automatic Road Extraction from Historical Maps using Deep Learning Techniques\n\n- [Istanbul_Dataset](https://github.com/TolgaBkm/Istanbul_Dataset) -\u003e segmentation on the Istanbul, Inria and Massachusetts datasets\n\n- [Road-Segmentation](https://github.com/ralph-elhaddad/Road-Segmentation) -\u003e Road segmentation on Satellite Images using CNN (U-Nets and FCN8) and Logistic Regression\n\n- [D-LinkNet](https://github.com/ShenweiXie/D-LinkNet) -\u003e 1st place solution in DeepGlobe Road Extraction Challenge\n\n- [PaRK-Detect](https://github.com/ShenweiXie/PaRK-Detect) -\u003e PaRK-Detect: Towards Efficient Multi-Task Satellite Imagery Road Extraction via Patch-Wise Keypoints Detection\n\n- [tile2net](https://github.com/VIDA-NYU/tile2net) -\u003e Mapping the walk: A scalable computer vision approach for generating sidewalk network datasets from aerial imagery\n\n- [sam_road](https://github.com/htcr/sam_road) -\u003e Segment Anything Model (SAM) for large-scale, vectorized road network extraction from aerial imagery.\n\n- [LRDNet](https://github.com/dyl96/LRDNet) -\u003e A Lightweight Road Detection Algorithm Based on Multiscale Convolutional Attention Network and Coupled Decoder Head\n\n- [Fine–Grained Extraction of Road Networks via Joint Learning of Connectivity and Segmentation](https://github.com/YXu556/RoadExtraction) -\u003e uses SpaceNet 3 dataset\n\n- [Satellite-Image-Road-Segmentation](https://github.com/aavek/Satellite-Image-Road-Segmentation) -\u003e Graph Reasoned Multi-Scale Road Segmentation in Remote Sensing Imagery [paper](https://ieeexplore.ieee.org/document/10281660)\n\n### Segmentation - Buildings \u0026 rooftops\n\n- [Road and Building Semantic Segmentation in Satellite Imagery](https://github.com/Paulymorphous/Road-Segmentation) uses U-Net on the Massachusetts Roads Dataset \u0026 keras\n\n- [find unauthorized constructions using aerial photography](https://medium.com/towards-artificial-intelligence/find-unauthorized-constructions-using-aerial-photography-and-deep-learning-with-code-part-2-b56ca80c8c99) -\u003e [Dataset creation](https://pub.towardsai.net/find-unauthorized-constructions-using-aerial-photography-and-deep-learning-with-code-part-1-6d3ca7ff6fa0)\n\n- [SRBuildSeg](https://github.com/xian1234/SRBuildSeg) -\u003e Making low-resolution satellite images reborn: a deep learning approach for super-resolution building extraction\n\n- [Building footprint detection with fastai on the challenging SpaceNet7 dataset](https://deeplearning.berlin/satellite%20imagery/computer%20vision/fastai/2021/02/17/Building-Detection-SpaceNet7.html) uses U-Net \u0026 fastai\n\n- [Pix2Pix-for-Semantic-Segmentation-of-Satellite-Images](https://github.com/A2Amir/Pix2Pix-for-Semantic-Segmentation-of-Satellite-Images) -\u003e using Pix2Pix GAN network to segment the building footprint from Satellite Images, uses tensorflow\n\n- [SpaceNetUnet](https://github.com/boggis30/SpaceNetUnet) -\u003e Baseline model is U-net like, applied to SpaceNet Vegas data, using Keras\n\n- [automated-building-detection](https://github.com/rodekruis/automated-building-detection) -\u003e Input: very-high-resolution (\u003c= 0.5 m/pixel) RGB satellite images. Output: buildings in vector format (geojson), to be used in digital map products. Built on top of robosat and robosat.pink.\n\n- [project_sunroof_india](https://github.com/AKASH2907/project_sunroof_india) -\u003e Analyzed Google Satellite images to generate a report on individual house rooftop's solar power potential, uses a range of classical computer vision techniques (e.g Canny Edge Detection) to segment the roofs\n\n- [JointNet-A-Common-Neural-Network-for-Road-and-Building-Extraction](https://github.com/ThomasWangWeiHong/JointNet-A-Common-Neural-Network-for-Road-and-Building-Extraction)\n\n- [Mapping Africa’s Buildings with Satellite Imagery: Google AI blog post](https://ai.googleblog.com/2021/07/mapping-africas-buildings-with.html). See the [open-buildings](https://sites.research.google/open-buildings/) dataset\n\n- [nz_convnet](https://github.com/weiji14/nz_convnet) -\u003e A U-net based ConvNet for New Zealand imagery to classify building outlines\n\n- [polycnn](https://github.com/Lydorn/polycnn) -\u003e End-to-End Learning of Polygons for Remote Sensing Image Classification\n\n- [spacenet_building_detection](https://github.com/motokimura/spacenet_building_detection) solution by [motokimura](https://github.com/motokimura) using Unet\n\n- [Vec2Instance](https://github.com/lakmalnd/Vec2Instance) -\u003e applied to the SpaceNet challenge AOI 2 (Vegas) building footprint dataset, tensorflow v1.12\n\n- [EarthquakeDamageDetection](https://github.com/JaneKravchenko/EarthquakeDamageDetection) -\u003e Buildings segmentation from satellite imagery and damage classification for each build, using Keras\n\n- [Semantic-segmentation repo by fuweifu-vtoo](https://github.com/fuweifu-vtoo/Semantic-segmentation) -\u003e uses pytorch and the [Massachusetts Buildings \u0026 Roads Datasets](https://www.cs.toronto.edu/~vmnih/data/)\n\n- [Extracting buildings and roads from AWS Open Data using Amazon SageMaker](https://aws.amazon.com/blogs/machine-learning/extracting-buildings-and-roads-from-aws-open-data-using-amazon-sagemaker/) -\u003e With [repo](https://github.com/aws-samples/aws-open-data-satellite-lidar-tutorial)\n\n- [TF-SegNet](https://github.com/mathildor/TF-SegNet) -\u003e AirNet is a segmentation network based on SegNet, but with some modifications\n\n- [rgb-footprint-extract](https://github.com/aatifjiwani/rgb-footprint-extract) -\u003e a Semantic Segmentation Network for Urban-Scale Building Footprint Extraction Using RGB Satellite Imagery, DeepLavV3+ module with a Dilated ResNet C42 backbone\n\n- [SpaceNetExploration](https://github.com/yangsiyu007/SpaceNetExploration) -\u003e A sample project demonstrating how to extract building footprints from satellite images using a semantic segmentation model. Data from the SpaceNet Challenge\n\n- [Rooftop-Instance-Segmentation](https://github.com/MasterSkepticista/Rooftop-Instance-Segmentation) -\u003e VGG-16, Instance Segmentation, uses the Airs dataset\n\n- [solar-farms-mapping](https://github.com/microsoft/solar-farms-mapping) -\u003e An Artificial Intelligence Dataset for Solar Energy Locations in India\n\n- [poultry-cafos](https://github.com/microsoft/poultry-cafos) -\u003e This repo contains code for detecting poultry barns from high-resolution aerial imagery and an accompanying dataset of predicted barns over the United States\n\n- [ssai-cnn](https://github.com/mitmul/ssai-cnn) -\u003e This is an implementation of Volodymyr Mnih's dissertation methods on his Massachusetts road \u0026 building dataset\n\n- [Remote-sensing-building-extraction-to-3D-model-using-Paddle-and-Grasshopper](https://github.com/Youssef-Harby/Remote-sensing-building-extraction-to-3D-model-using-Paddle-and-Grasshopper)\n\n- [segmentation-enhanced-resunet](https://github.com/tranleanh/segmentation-enhanced-resunet) -\u003e Urban building extraction in Daejeon region using Modified Residual U-Net (Modified ResUnet) and applying post-processing\n\n- [Mask RCNN for Spacenet Off Nadir Building Detection](https://github.com/ashnair1/Mask-RCNN-for-Off-Nadir-Building-Detection)\n\n- [GRSL_BFE_MA](https://github.com/jiankang1991/GRSL_BFE_MA) -\u003e Deep Learning-based Building Footprint Extraction with Missing Annotations using a novel loss function\n\n- [FER-CNN](https://github.com/runnergirl13/FER-CNN) -\u003e Detection, Classification and Boundary Regularization of Buildings in Satellite Imagery Using Faster Edge Region Convolutional Neural Networks\n\n- [UNET-Image-Segmentation-Satellite-Picture](https://github.com/rwie1and/UNET-Image-Segmentation-Satellite-Pictures) -\u003e Unet to predict roof tops on Crowed AI Mapping dataset, uses keras\n\n- [Vector-Map-Generation-from-Aerial-Imagery-using-Deep-Learning-GeoSpatial-UNET](https://github.com/ManishSahu53/Vector-Map-Generation-from-Aerial-Imagery-using-Deep-Learning-GeoSpatial-UNET) -\u003e applied to geo-referenced images which are very large size \u003e 10k x 10k pixels\n\n- [building-footprint-segmentation](https://github.com/fuzailpalnak/building-footprint-segmentation) -\u003e pip installable library to train building footprint segmentation on satellite and aerial imagery, applied to Massachusetts Buildings Dataset and Inria Aerial Image Labeling Dataset\n\n- [SemSegBuildings](https://github.com/SharpestProjects/SemSegBuildings) -\u003e Project using fast.ai framework for semantic segmentation on Inria building segmentation dataset\n\n- [FCNN-example](https://github.com/emredog/FCNN-example) -\u003e overfit to a given single image to detect houses\n\n- [SAT2LOD2](https://github.com/gdaosu/lod2buildingmodel) -\u003e an open-source, python-based GUI-enabled software that takes the satellite images as inputs and returns LoD2 building models as outputs\n\n- [SatFootprint](https://github.com/PriyanK7n/SatFootprint) -\u003e building segmentation on the Spacenet 7 dataset\n\n- [Building-Detection](https://github.com/EL-BID/Building-Detection) -\u003e Raster Vision experiment to train a model to detect buildings from satellite imagery in three cities in Latin America\n\n- [Multi-building-tracker](https://github.com/sebasmos/Multi-building-tracker) -\u003e Multi-target building tracker for satellite images using deep learning\n\n- [Boundary Enhancement Semantic Segmentation for Building Extraction](https://github.com/hin1115/BEmodule-Satellite-Building-Segmentation)\n\n- [keras code for binary semantic segmentation](https://github.com/loveswine/UNet_keras_for_RSimage)\n\n- [Spacenet-Building-Detection](https://github.com/IdanC1s2/Spacenet-Building-Detection)\n\n- [LGPNet-BCD](https://github.com/TongfeiLiu/LGPNet-BCD) -\u003e Building Change Detection for VHR Remote Sensing Images via Local-Global Pyramid Network and Cross-Task Transfer Learning Strategy\n\n- [MTL_homoscedastic_SRB](https://github.com/burakekim/MTL_homoscedastic_SRB) -\u003e A Multi-Task Deep Learning Framework for Building Footprint Segmentation\n\n- [UNet_CNN](https://github.com/Inamdarpushkar/UNet_CNN) -\u003e UNet model to segment building coverage in Boston using Remote sensing data, uses keras\n\n- [FDANet](https://github.com/daifeng2016/FDANet) -\u003e Full-Level Domain Adaptation for Building Extraction in Very-High-Resolution Optical Remote-Sensing Images\n\n- [CBRNet](https://github.com/HaonanGuo/CBRNet) -\u003e A Coarse-to-fine Boundary Refinement Network for Building Extraction from Remote Sensing Imagery\n\n- [ASLNet](https://github.com/ggsDing/ASLNet) -\u003e Adversarial Shape Learning for Building Extraction in VHR Remote Sensing Images\n\n- [BRRNet](https://github.com/wangyi111/Building-Extraction) -\u003e A Fully Convolutional Neural Network for Automatic Building Extraction From High-Resolution Remote Sensing Images\n\n- [Multi-Scale-Filtering-Building-Index](https://github.com/ThomasWangWeiHong/Multi-Scale-Filtering-Building-Index) -\u003e A Multi - Scale Filtering Building Index for Building Extraction in Very High - Resolution Satellite Imagery\n\n- [Models for Remote Sensing](https://github.com/bohaohuang/mrs) -\u003e long list of unets etc applied to building detection\n\n- [boundary_loss_for_remote_sensing](https://github.com/yiskw713/boundary_loss_for_remote_sensing) -\u003e Boundary Loss for Remote Sensing Imagery Semantic Segmentation\n\n- [Open Cities AI Challenge](https://www.drivendata.org/competitions/60/building-segmentation-disaster-resilience/) -\u003e Segmenting Buildings for Disaster Resilience. Winning solutions [on Github](https://github.com/drivendataorg/open-cities-ai-challenge/)\n\n- [MAPNet](https://github.com/lehaifeng/MAPNet) -\u003e Multi Attending Path Neural Network for Building Footprint Extraction from Remote Sensed Imagery\n\n- [dual-hrnet](https://github.com/SIAnalytics/dual-hrnet) -\u003e localizing buildings and classifying their damage level\n\n- [ESFNet](https://github.com/mrluin/ESFNet-Pytorch) -\u003e Efficient Network for Building Extraction from High-Resolution Aerial Images\n\n- [rooftop-detection-python](https://github.com/sayonpalit/rooftop-detection-python) -\u003e Detect Rooftops from low resolution satellite images and calculate area for cultivation and solar panel installment using classical computer vision techniques\n\n- [keras_segmentation_models](https://github.com/sajmonogy/keras_segmentation_models) -\u003e Using Open Vector-Based Spatial Data to Create Semantic Datasets for Building Segmentation for Raster Data\n\n- [CVCMFFNet](https://github.com/Jiankun-chen/CVCMFFNet-master) -\u003e Complex-Valued Convolutional and Multifeature Fusion Network for Building Semantic Segmentation of InSAR Images\n\n- [STEB-UNet](https://github.com/BrightGuo048/STEB-UNet) -\u003e A Swin Transformer-Based Encoding Booster Integrated in U-Shaped Network for Building Extraction\n\n- [dfc2020_baseline](https://github.com/lukasliebel/dfc2020_baseline) -\u003e Baseline solution for the IEEE GRSS Data Fusion Contest 2020. Predict land cover labels from Sentinel-1 and Sentinel-2 imagery\n\n- [Fusing multiple segmentation models based on different datasets into a single edge-deployable model](https://github.com/markusmeingast/Satellite-Classifier) -\u003e roof, car \u0026 road segmentation\n\n- [ground-truth-gan-segmentation](https://github.com/zakariamejdoul/ground-truth-gan-segmentation) -\u003e use Pix2Pix to segment the footprint of a building. The dataset used is AIRS\n\n- [UNICEF-Giga_Sudan](https://github.com/Kamal-Eldin/UNICEF-Giga_Sudan) -\u003e Detecting school lots from satellite imagery in Southern Sudan using a UNET segmentation model\n\n- [building_footprint_extraction](https://github.com/shubhamgoel27/building_footprint_extraction) -\u003e The project retrieves satellite imagery from Google and performs building footprint extraction using a U-Net.\n\n- [projectRegularization](https://github.com/zorzi-s/projectRegularization) -\u003e Regularization of building boundaries in satellite images using adversarial and regularized losses\n\n- [PolyWorldPretrainedNetwork](https://github.com/zorzi-s/PolyWorldPretrainedNetwork) -\u003e Polygonal Building Extraction with Graph Neural Networks in Satellite Images\n\n- [dl_image_segmentation](https://github.com/harry-gibson/dl_image_segmentation) -\u003e Uncertainty-Aware Interpretable Deep Learning for Slum Mapping and Monitoring. Uses SHAP\n\n- [UBC-dataset](https://github.com/AICyberTeam/UBC-dataset) -\u003e a dataset for building detection and classification from very high-resolution satellite imagery with the focus on object-level interpretation of individual buildings\n\n- [UNetFormer](https://github.com/WangLibo1995/GeoSeg) -\u003e A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery\n\n- [BES-Net](https://github.com/FlyC235/BESNet) -\u003e Boundary Enhancing Semantic Context Network for High-Resolution Image Semantic Segmentation. Applied to Vaihingen and Potsdam datasets\n\n- [CVNet](https://github.com/xzq-njust/CVNet) -\u003e Contour Vibration Network for Building Extraction\n\n- [CFENet](https://github.com/djzgroup/CFENet) -\u003e A Context Feature Enhancement Network for Building Extraction from High-Resolution Remote Sensing Imagery\n\n- [HiSup](https://github.com/SarahwXU/HiSup) -\u003e Accurate Polygonal Mapping of Buildings in Satellite Imagery\n\n- [BuildingExtraction](https://github.com/KyanChen/BuildingExtraction) -\u003e Building Extraction from Remote Sensing Images with Sparse Token Transformers\n\n- [CrossGeoNet](https://github.com/lqycrystal/coseg_building) -\u003e A Framework for Building Footprint Generation of Label-Scarce Geographical Regions\n\n- [AFM_building](https://github.com/lqycrystal/AFM_building) -\u003e Building Footprint Generation Through Convolutional Neural Networks With Attraction Field Representation\n\n- [RAMP (Replicable AI for MicroPlanning)](https://github.com/devglobalpartners/ramp-code) -\u003e building detection in low and middle income countries\n\n- [Building-instance-segmentation](https://github.com/yuanqinglie/Building-instance-segmentation-combining-anchor-free-detectors-and-multi-modal-feature-fusion) -\u003e Multi-Modal Feature Fusion Network with Adaptive Center Point Detector for Building Instance Extraction\n\n- [CGSANet](https://github.com/MrChen18/CGSANet) -\u003e A Contour-Guided and Local Structure-Aware Encoder–Decoder Network for Accurate Building Extraction From Very High-Resolution Remote Sensing Imagery\n\n- [building-footprints-update](https://github.com/wangzehui20/building-footprints-update) -\u003e Learning Color Distributions from Bitemporal Remote Sensing Images to Update Existing Building Footprints\n\n- [RAMP](https://rampml.global/) -\u003e model and buildings dataset to support a wide variety of humanitarian use cases\n\n- [Thesis_Semantic_Image_Segmentation_on_Satellite_Imagery_using_UNets](https://github.com/rinkwitz/Thesis_Semantic_Image_Segmentation_on_Satellite_Imagery_using_UNets) -\u003e This master thesis aims to perform semantic segmentation of buildings on satellite images from the SpaceNet challenge 1 dataset using the U-Net architecture\n\n- [HD-Net](https://github.com/danfenghong/ISPRS_HD-Net) -\u003e High-resolution decoupled network for building footprint extraction via deeply supervised body and boundary decomposition\n\n- [RoofSense](https://github.com/DimitrisMantas/RoofSense/tree/master) -\u003e A novel deep learning solution for the automatic roofing material classification of the Dutch building stock using aerial imagery and laser scanning data fusion\n\n- [IBS-AQSNet](https://github.com/zhilyzhang/IBS-AQSNet) -\u003e Enhanced Automated Quality Assessment Network for Interactive Building Segmentation in High-Resolution Remote Sensing Imagery\n\n- [DeepMAO](https://github.com/Sumanth181099/DeepMAO) -\u003e Deep Multi-scale Aware Overcomplete Network for Building Segmentation in Satellite Imagery\n\n- [CMGFNet-Building_Extraction](https://github.com/hamidreza2015/CMGFNet-Building_Extraction) -\u003e Deep Learning Code for Building Extraction from very high resolution (VHR) remote sensing images\n\n### Segmentation - Solar panels\n\n- [Deep-Learning-for-Solar-Panel-Recognition](https://github.com/saizk/Deep-Learning-for-Solar-Panel-Recognition) -\u003e using both object detection with Yolov5 and Unet segmentation\n\n- [DeepSolar](https://github.com/wangzhecheng/DeepSolar) -\u003e A Machine Learning Framework to Efficiently Construct a Solar Deployment Database in the United States. [Dataset on kaggle](https://www.kaggle.com/datasets/tunguz/deep-solar-dataset), actually used a CNN for classification and segmentation is obtained by applying a threshold to the activation map. Original code is tf1 but [tf2/kers](https://github.com/aidan-fitz/deepsolar-v2) and a [pytorch implementation](https://github.com/wangzhecheng/deepsolar_pytorch) are available. Also checkout [Visualizations and in-depth analysis .. of the factors that can explain the adoption of solar energy in ..  Virginia]\n  \n- [hyperion_solar_net](https://github.com/fvergaracontesse/hyperion_solar_net) -\u003e trained classificaton \u0026 segmentation models on RGB imagery from Google Maps\n\n- [3D-PV-Locator](https://github.com/kdmayer/3D-PV-Locator) -\u003e Large-scale detection of rooftop-mounted photovoltaic systems in 3D\n\n- [PV_Pipeline](https://github.com/kdmayer/PV_Pipeline) -\u003e DeepSolar for Germany\n\n- [solar-panels-detection](https://github.com/dbaofd/solar-panels-detection) -\u003e using SegNet, Fast SCNN \u0026 ResNet\n\n- [predict_pv_yield](https://github.com/openclimatefix/predict_pv_yield) -\u003e Using optical flow \u0026 machine learning to predict PV yield\n\n- [Large-scale-solar-plant-monitoring](https://github.com/osmarluiz/Large-scale-solar-plant-monitoring) -\u003e Remote Sensing for Monitoring of Photovoltaic Power Plants in Brazil Using Deep Semantic Segmentation\n\n- [Panel-Segmentation](https://github.com/NREL/Panel-Segmentation) -\u003e Determine the presence of a solar array in the satellite image (boolean True/False), using a VGG16 classification model\n\n- [Roofpedia](https://github.com/ualsg/Roofpedia) -\u003e an open registry of green roofs and solar roofs across the globe identified by Roofpedia through deep learning\n\n- [Predicting the Solar Potential of Rooftops using Image Segmentation and Structured Data](https://medium.com/nam-r/predicting-the-solar-potential-of-rooftops-using-image-segmentation-and-structured-data-61198c39d57c) Medium article, using 20cm imagery \u0026 Unet\n\n- [solar-pv-global-inventory](https://github.com/Lkruitwagen/solar-pv-global-inventory)\n\n- [remote-sensing-solar-pv](https://github.com/Lkruitwagen/remote-sensing-solar-pv) -\u003e A repository for sharing progress on the automated detection of solar PV arrays in sentinel-2 remote sensing imagery\n\n- [solar-panel-segmentation)](https://github.com/gabrieltseng/solar-panel-segmentation) -\u003e Finding solar panels using USGS satellite imagery\n\n- [solar_seg](https://github.com/tcapelle/solar_seg) -\u003e Solar segmentation of PV modules (sub elements of panels) using drone images and fast.ai\n\n- [solar_plant_detection](https://github.com/Amirmoradi94/solar_plant_detection) -\u003e boundary extraction of Photovoltaic (PV) plants using Mask RCNN and Amir dataset\n\n- [SolarDetection](https://github.com/A-Stangeland/SolarDetection) -\u003e unet on satellite image from the USA and France\n\n- [adopptrs](https://github.com/francois-rozet/adopptrs) -\u003e Automatic Detection Of Photovoltaic Panels Through Remote Sensing using unet \u0026 pytorch\n\n- [solar-panel-locator](https://github.com/TorrBorr/solar-panel-locator) -\u003e the number of solar panel pixels was only ~0.2% of the total pixels in the dataset, so solar panel data was upsampled to account for the class imbalance\n\n- [projects-solar-panel-detection](https://github.com/top-on/projects-solar-panel-detection) -\u003e List of project to detect solar panels from aerial/satellite images\n\n- [Satellite_ComputerVision](https://github.com/mjevans26/Satellite_ComputerVision) -\u003e UNET to detect solar arrays from Sentinel-2 data, using Google Earth Engine and Tensorflow. Also covers parking lot detection\n\n- [photovoltaic-detection](https://github.com/riccardocadei/photovoltaic-detection) -\u003e Detecting available rooftop area from satellite images to install photovoltaic panels\n\n- [Solar_UNet](https://github.com/mjevans26/Solar_UNet) -\u003e U-Net models delineating solar arrays in Sentinel-2 imagery\n\n- [SolarDetection-solafune](https://github.com/bit-guber/SolarDetection-solafune) -\u003e Solar Panel Detection Using Sentinel-2 for the Solafune Competition\n\n- [A Comparative Evaluation of Deep Learning Techniques for Photovoltaic Panel Detection from Aerial Images](https://github.com/links-ads/access-solar-panels)\n\n- [UCSD_MLBootcamp_Capstone](https://github.com/FederCO23/UCSD_MLBootcamp_Capstone) -\u003e Automatic Detection of Photovoltaic Power Stations Using Satellite Imagery and Deep Learning (Sentinel 2)\n\n### Segmentation - Ships \u0026 vessels\n\n- [Universal-segmentation-baseline-Kaggle-Airbus-Ship-Detection](https://github.com/OniroAI/Universal-segmentation-baseline-Kaggle-Airbus-Ship-Detection) -\u003e Kaggle Airbus Ship Detection Challenge - bronze medal solution\n\n- [Airbus-Ship-Segmentation](https://github.com/TheXirex/Airbus-Ship-Segmentation) -\u003e unet\n\n- [contrastive_SSL_ship_detection](https://github.com/alina2204/contrastive_SSL_ship_detection) -\u003e Contrastive self supervised learning for ship detection in Sentinel 2 images\n\n- [airbus-ship-detection](https://github.com/odessitua/airbus-ship-detection) -\u003e using DeepLabV3+\n\n- [Unet with web-application applied to Airbus ships](https://github.com/glibesyck/ImageSegmentation)\n\n### Segmentation - Other manmade\n\n- [Aarsh2001/ML_Challenge_NRSC](https://github.com/Aarsh2001/ML_Challenge_NRSC) -\u003e Electrical Substation detection\n\n- [electrical_substation_detection](https://github.com/thisishardik/electrical_substation_detection)\n\n- [MCAN-OilSpillDetection](https://github.com/liyongqingupc/MCAN-OilSpillDetection) -\u003e Oil Spill Detection with A Multiscale Conditional Adversarial Network under Small Data Training\n\n- [mining-detector](https://github.com/earthrise-media/mining-detector) -\u003e detection of artisanal gold mines in Sentinel-2 satellite imagery for [Amazon Mining Watch](https://amazonminingwatch.org/). Also covers clandestine airstrips\n\n- [EG-UNet](https://github.com/tist0bsc/EG-UNet) Deep Feature Enhancement Method for Land Cover With Irregular and Sparse Spatial Distribution Features: A Case Study on Open-Pit Mining\n\n- [plastics](https://github.com/earthrise-media/plastics) -\u003e Detecting and Monitoring Plastic Waste Aggregations in Sentinel-2 Imagery\n\n- [MADOS](https://github.com/gkakogeorgiou/mados) -\u003e Detecting Marine Pollutants and Sea Surface Features with Deep Learning in Sentinel-2 Imagery on the MADOS dataset\n\n- [SADMA](https://github.com/sheikhazhanmohammed/SADMA) -\u003e Residual Attention UNet on MARIDA: Marine Debris Archive is a marine debris-oriented dataset on Sentinel-2 satellite images\n\n- [MAP-Mapper](https://github.com/CoDIS-Lab/MAP-Mapper) -\u003e Marine Plastic Mapper is a tool for assessing marine macro-plastic density to identify plastic hotspots, underpinned by the MARIDA dataset.\n\n- [substation-seg](https://github.com/Lindsay-Lab/substation-seg) -\u003e segmenting substations in Sentinel 2 satellite imagery\n\n### Panoptic segmentation\n\n- [Things and stuff or how remote sensing could benefit from panoptic segmentation](https://softwaremill.com/things-and-stuff-or-how-remote-sensing-could-benefit-from-panoptic-segmentation/)\n\n- [utae-paps](https://github.com/VSainteuf/utae-paps) -\u003e PyTorch implementation of U-TAE and PaPs for satellite image time series panoptic segmentation\n\n- [pastis-benchmark](https://github.com/VSainteuf/pastis-benchmark)\n\n- [Panoptic-Generator](https://github.com/abilius-app/Panoptic-Generator) -\u003e This module converts GIS data into panoptic segmentation tiles\n\n- [BSB-Aerial-Dataset](https://github.com/osmarluiz/BSB-Aerial-Dataset) -\u003e an example on how to use Detectron2's Panoptic-FPN in the BSB Aerial Dataset\n\n### Segmentation - Miscellaneous\n\n- [seg-eval](https://github.com/itracasa/seg-eval) -\u003e SegEval is a Python library that provides tools for evaluating semantic segmentation models. Generate evaluation regions and to analyze segmentation results within them.\n\n- [awesome-satellite-images-segmentation](https://github.com/mrgloom/awesome-semantic-segmentation#satellite-images-segmentation)\n\n- [Satellite Image Segmentation: a Workflow with U-Net](https://medium.com/vooban-ai/satellite-image-segmentation-a-workflow-with-u-net-7ff992b2a56e) is a decent intro article\n\n- [mmsegmentation](https://github.com/open-mmlab/mmsegmentation) -\u003e Semantic Segmentation Toolbox with support for many remote sensing datasets including LoveDA, Potsdam, Vaihingen \u0026 iSAID\n\n- [segmentation_gym](https://github.com/Doodleverse/segmentation_gym) -\u003e A neural gym for training deep learning models to carry out geoscientific image segmentation\n\n- [Using a U-Net for image segmentation, blending predicted patches smoothly is a must to please the human eye](https://github.com/Vooban/Smoothly-Blend-Image-Patches) -\u003e python code to blend predicted patches smoothly. See [Satellite-Image-Segmentation-with-Smooth-Blending](https://github.com/MaitrySinha21/Satellite-Image-Segmentation-with-Smooth-Blending)\n\n- [DCA](https://github.com/Luffy03/DCA) -\u003e Deep Covariance Alignment for Domain Adaptive Remote Sensing Image Segmentation\n\n- [SCAttNet](https://github.com/lehaifeng/SCAttNet) -\u003e Semantic Segmentation Network with Spatial and Channel Attention Mechanism\n\n- [unetseg](https://github.com/dymaxionlabs/unetseg) -\u003e A set of classes and CLI tools for training a semantic segmentation model based on the U-Net architecture, using Tensorflow and Keras. This implementation is tuned specifically for satellite imagery and other geospatial raster data\n\n- [Semantic Segmentation of Satellite Imagery using U-Net \u0026 fast.ai](https://medium.com/dataseries/image-semantic-segmentation-of-satellite-imagery-using-u-net-e99ae13cf464) -\u003e with [repo](https://github.com/raoofnaushad/Image-Semantic-Segmentation-of-Satellite-Imagery-using-U-Net.)\n\n- [clusternet_segmentation](https://github.com/zhygallo/clusternet_segmentation) -\u003e Unsupervised Segmentation by applying K-Means clustering to the features generated by Neural Network\n\n- [Efficient-Transformer](https://github.com/zyxu1996/Efficient-Transformer) -\u003e Efficient Transformer for Remote Sensing Image Segmentation\n\n- [weakly_supervised](https://github.com/LobellLab/weakly_supervised) -\u003e Weakly Supervised Deep Learning for Segmentation of Remote Sensing Imagery\n\n- [HRCNet-High-Resolution-Context-Extraction-Network](https://github.com/zyxu1996/HRCNet-High-Resolution-Context-Extraction-Network) -\u003e High-Resolution Context Extraction Network for Semantic Segmentation of Remote Sensing Images\n\n- [Semantic segmentation of SAR images using a self supervised technique](https://github.com/cattale93/pytorch_self_supervised_learning)\n\n- [satellite-segmentation-pytorch](https://github.com/obravo7/satellite-segmentation-pytorch) -\u003e explores a wide variety of image augmentations to increase training dataset size\n\n- [Spectralformer](https://github.com/danfenghong/IEEE_TGRS_SpectralFormer) -\u003e Rethinking hyperspectral image classification with transformers\n\n- [Unsupervised Segmentation of Hyperspectral Remote Sensing Images with Superpixels](https://github.com/mpBarbato/Unsupervised-Segmentation-of-Hyperspectral-Remote-Sensing-Images-with-Superpixels)\n\n- [Semantic-Segmentation-with-Sparse-Labels](https://github.com/Hua-YS/Semantic-Segmentation-with-Sparse-Labels)\n\n- [SNDF](https://github.com/mi18/SNDF) -\u003e Superpixel-enhanced deep neural forest for remote sensing image semantic segmentation\n\n- [Satellite-Image-Classification](https://github.com/yxian29/Satellite-Image-Classification) -\u003e using random forest or support vector machines (SVM) and sklearn\n\n- [dynamic-rs-segmentation](https://github.com/keillernogueira/dynamic-rs-segmentation) -\u003e Dynamic Multi-Context Segmentation of Remote Sensing Images based on Convolutional Networks\n\n- [segmentation_models.pytorch](https://github.com/qubvel/segmentation_models.pytorch) -\u003e Segmentation models with pretrained backbones, has been used in multiple winning solutions to remote sensing competitions\n\n- [SSRN](https://github.com/zilongzhong/SSRN) -\u003e Spectral-Spatial Residual Network for Hyperspectral Image Classification: A 3-D Deep Learning Framework\n\n- [SO-DNN](https://github.com/PanXinZebra/SO-DNN) -\u003e Simplified object-based deep neural network for very high resolution remote sensing image classification\n\n- [SANet](https://github.com/mrluin/SANet-PyTorch) -\u003e Scale-Aware Network for Semantic Segmentation of High-Resolution Aerial Images\n\n- [aerial-segmentation](https://github.com/alpemek/aerial-segmentation) -\u003e Learning Aerial Image Segmentation from Online Maps\n\n- [IterativeSegmentation](https://github.com/gaudetcj/IterativeSegmentation) -\u003e Recurrent Neural Networks to Correct Satellite Image Classification Maps\n\n- [Detectron2 FPN + PointRend Model for amazing Satellite Image Segmentation](https://affine.medium.com/detectron2-fpn-pointrend-model-for-amazing-satellite-image-segmentation-183456063e15) -\u003e 15% increase in accuracy when compared to the U-Net model\n\n- [HybridSN](https://github.com/purbayankar/HybridSN-pytorch) -\u003e Exploring 3D-2D CNN Feature Hierarchy for Hyperspectral Image Classification\n\n- [TNNLS_2022_X-GPN](https://github.com/B-Xi/TNNLS_2022_X-GPN) -\u003e Semisupervised Cross-scale Graph Prototypical Network for Hyperspectral Image Classification\n\n- [singleSceneSemSegTgrs2022](https://github.com/sudipansaha/singleSceneSemSegTgrs2022) -\u003e Unsupervised Single-Scene Semantic Segmentation for Earth Observation\n\n- [A-Fast-and-Compact-3-D-CNN-for-HSIC](https://github.com/mahmad00/A-Fast-and-Compact-3-D-CNN-for-HSIC) -\u003e A Fast and Compact 3-D CNN for Hyperspectral Image Classification\n\n- [HSNRS](https://github.com/Walkerlikesfish/HSNRS) -\u003e  Hourglass-ShapeNetwork Based Semantic Segmentation for High Resolution Aerial Imagery\n\n- [GiGCN](https://github.com/ShuGuoJ/GiGCN) -\u003e Graph-in-Graph Convolutional Network for Hyperspectral Image Classification\n\n- [SSAN](https://github.com/EtPan/SSAN) -\u003e Spectral-Spatial Attention Networks for Hyperspectral Image Classification\n\n- [drone-images-semantic-segmentation](https://github.com/ayushdabra/drone-images-semantic-segmentation) -\u003e Multiclass Semantic Segmentation of Aerial Drone Images Using Deep Learning\n\n- [Satellite-Image-Segmentation-with-Smooth-Blending](https://github.com/MaitrySinha21/Satellite-Image-Segmentation-with-Smooth-Blending) -\u003e uses [Smoothly-Blend-Image-Patches](https://github.com/Vooban/Smoothly-Blend-Image-Patches)\n\n- [BayesianUNet](https://github.com/tha-santacruz/BayesianUNet) -\u003e Pytorch Bayesian UNet model for segmentation and uncertainty prediction, applied to the Potsdam Dataset\n\n- [RAANet](https://github.com/Lrr0213/RAANet) -\u003e A Residual ASPP with Attention Framework for Semantic Segmentation of High-Resolution Remote Sensing Images\n\n- [wheelRuts_semanticSegmentation](https://github.com/SmartForest-no/wheelRuts_semanticSegmentation) -\u003e Mapping wheel-ruts from timber harvesting operations using deep learning techniques in drone imagery\n\n- [LWN-for-UAVRSI](https://github.com/syliudf/LWN-for-UAVRSI) -\u003e Light-Weight Semantic Segmentation Network for UAV Remote Sensing Images, applied to Vaihingen, UAVid and UDD6 datasets\n\n- [hypernet](https://github.com/ESA-PhiLab/hypernet) -\u003e library which implements hyperspectral image (HSI) segmentation\n\n- [ST-UNet](https://github.com/XinnHe/ST-UNet) -\u003e Swin Transformer Embedding UNet for Remote Sensing Image Semantic Segmentation\n\n- [EDFT](https://github.com/h1063135843/EDFT) -\u003e Efficient Depth Fusion Transformer for Aerial Image Semantic Segmentation\n\n- [WiCoNet](https://github.com/ggsDing/WiCoNet) -\u003e Looking Outside the Window: Wide-Context Transformer for the Semantic Segmentation of High-Resolution Remote Sensing Images\n\n- [CRGNet](https://github.com/YonghaoXu/CRGNet) -\u003e Consistency-Regularized Region-Growing Network for Semantic Segmentation of Urban Scenes with Point-Level Annotations\n\n- [SA-UNet](https://github.com/Yancccccc/SA-UNet) -\u003e Improved U-Net Remote Sensing Classification Algorithm Fusing Attention and Multiscale Features\n\n- [MANet](https://github.com/lironui/Multi-Attention-Network) -\u003e Multi-Attention-Network for Semantic Segmentation of Fine Resolution Remote Sensing Images\n\n- [BANet](https://github.com/lironui/BANet) -\u003e Transformer Meets Convolution: A Bilateral Awareness Network for Semantic Segmentation of Very Fine Resolution Urban Scene Images\n\n- [MACU-Net](https://github.com/lironui/MACU-Net) -\u003e MACU-Net for Semantic Segmentation of Fine-Resolution Remotely Sensed Images\n\n- [DNAS](https://github.com/faye0078/DNAS) -\u003e Decoupling Neural Architecture Search for High-Resolution Remote Sensing Image Semantic Segmentation\n\n- [A2-FPN](https://github.com/lironui/A2-FPN) -\u003e A2-FPN for Semantic Segmentation of Fine-Resolution Remotely Sensed Images\n\n- [MAResU-Net](https://github.com/lironui/MAResU-Net) -\u003e Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images\n\n- [ml_segmentation](https://github.com/dgriffiths3/ml_segmentation) -\u003e semantic segmentation of buildings using Random Forest, Support Vector Machine (SVM) \u0026 Gradient Boosting Classifier (GBC)\n\n- [RSEN](https://github.com/YonghaoXu/RSEN) -\u003e Robust Self-Ensembling Network for Hyperspectral Image Classification\n\n- [MSNet](https://github.com/taochx/MSNet) -\u003e multispectral semantic segmentation network for remote sensing images\n\n- [k-textures](https://zenodo.org/record/6359859#.Yytt6OzMK3I) -\u003e K-textures, a self-supervised hard clustering deep learning algorithm for satellite image segmentation\n\n- [Swin-Transformer-Semantic-Segmentation](https://github.com/koechslin/Swin-Transformer-Semantic-Segmentation) -\u003e Satellite Image Semantic Segmentation\n\n- [UDA_for_RS](https://github.com/Levantespot/UDA_for_RS) -\u003e Unsupervised Domain Adaptation for Remote Sensing Semantic Segmentation with Transformer\n\n- [A-3D-CNN-AM-DSC-model-for-hyperspectral-image-classification](https://github.com/hahatongxue/A-3D-CNN-AM-DSC-model-for-hyperspectral-image-classification) -\u003e Attention Mechanism and Depthwise Separable Convolution Aided 3DCNN for Hyperspectral Remote Sensing Image Classification\n\n- [contrastive-distillation](https://github.com/edornd/contrastive-distillation) -\u003e  A Contrastive Distillation Approach for Incremental Semantic Segmentation in Aerial Images\n\n- [SegForestNet](https://github.com/gritzner/SegForestNet) -\u003e SegForestNet: Spatial-Partitioning-Based Aerial Image Segmentation\n\n- [MFVNet](https://github.com/weichenrs/MFVNet) -\u003e MFVNet: Deep Adaptive Fusion Network with Multiple Field-of-Views for Remote Sensing Image Semantic Segmentation\n\n- [Wildebeest-UNet](https://github.com/zijing-w/Wildebeest-UNet) -\u003e detecting wildebeest and zebras in Serengeti-Mara ecosystem from very-high-resolution satellite imagery\n\n- [segment-anything-eo](https://github.com/aliaksandr960/segment-anything-eo) -\u003e Earth observation tools for Meta AI Segment Anything (SAM - Segment Anything Model)\n\n- [HR-Image-classification_SDF2N](https://github.com/SicongLiuRS/HR-Image-classification_SDF2N) -\u003e A Shallow-to-Deep Feature Fusion Network for VHR Remote Sensing Image Classification\n\n- [sink-seg](https://github.com/mvrl/sink-seg) -\u003e Automatic Segmentation of Sinkholes Using a Convolutional Neural Network\n\n- [Tiling and Stitching Segmentation Output for Remote Sensing: Basic Challenges and Recommendations](https://arxiv.org/abs/1805.12219)\n\n- [EMRT](https://github.com/peach-xiao/EMRT) -\u003e Enhancing Multiscale Representations With Transformer for Remote Sensing Image Semantic Segmentation\n\n- [UDA_for_RS](https://github.com/Levantespot/UDA_for_RS) -\u003e Unsupervised Domain Adaptation for Remote Sensing Semantic Segmentation with Transformer\n\n- [CMTFNet](https://github.com/DrWuHonglin/CMTFNet) -\u003e CMTFNet: CNN and Multiscale Transformer Fusion Network for Remote Sensing Image Semantic Segmentation\n\n- [CM-UNet](https://github.com/XiaoBuL/CM-UNet) -\u003e Hybrid CNN-Mamba UNet for Remote Sensing Image Semantic Segmentation\n\n- [Using Stable Diffusion to Improve Image Segmentation Models](https://medium.com/edge-analytics/using-stable-diffusion-to-improve-image-segmentation-models-1e99c25acbf) -\u003e Augmenting Data with Stable Diffusion\n\n- [SSRS](https://github.com/sstary/SSRS) -\u003e Semantic Segmentation for Remote Sensing, multiple networks implemented\n\n- [BIOSCANN](https://github.com/BiodiversityLab/bioscann) -\u003e BIOdiversity Segmentation and Classification with Artificial Neural Networks\n\n- [ResUNet-a](https://github.com/feevos/resuneta) -\u003e a deep learning framework for semantic segmentation of remotely sensed data\n\n- [SSG2](https://github.com/feevos/ssg2) -\u003e A New Modelling Paradigm for Semantic Segmentation\n\n- [DBFNet](https://github.com/Luffy03/DBFNet) -\u003e Deep Bilateral Filtering Network for Point-Supervised Semantic Segmentation in Remote Sensing Images [IEEE](https://ieeexplore.ieee.org/document/9961229) paper\n\n- [PGNet](https://github.com/Fhujinwu/PGNet) -\u003e PGNet: Positioning Guidance Network for Semantic Segmentation of Very-High-Resolution Remote Sensing Images [paper](https://www.mdpi.com/2072-4292/14/17/4219)\n\n- [ASD](https://github.com/Jingtao-Li-CVer/ASD) -\u003e Anomaly Segmentation for High-Resolution Remote Sensing Images Based on Pixel Descriptors (AAAI2023) [download link](https://ojs.aaai.org/index.php/AAAI/article/view/25563/25335)\n\n- [u-nets-implementation](https://github.com/Aadit3003/u-nets-implementation) -\u003e Semantic-Segmentation-with-U-Nets\n\n- [SDM](https://github.com/caoql98/SDM) -\u003e Scale-aware Detailed Matching for Few-Shot Aerial Image Semantic Segmentation\n\n- [Transferability-Remote-Sensing](https://github.com/GDAOSU/Transferability-Remote-Sensing) -\u003e On the Transferability of Learning Models for Semantic Segmentation for Remote Sensing Data\n\n- [data-centric-satellite-segmentation](https://github.com/microsoft/data-centric-satellite-segmentation) -\u003e Contains implementations of data-centric approaches for improving semantic segmentation on satellite imagery, from Microsoft\n\n- [HSLabeling](https://github.com/linjiaxing99/HSLabeling) -\u003e Towards Efficient Labeling for Large-scale Remote Sensing Image Segmentation with Hybrid Sparse Labeling\n\n#\n## Instance segmentation\n\nIn instance segmentation, each individual 'instance' of a segmented area is given a unique lable. For detection of very small objects this may a good approach, but it can struggle seperating individual objects that are closely spaced.\n\n- [Mask_RCNN](https://github.com/matterport/Mask_RCNN) generates bounding boxes and segmentation masks for each instance of an object in the image. It is very commonly used for instance segmentation \u0026 object detection\n\n- [Instance segmentation of center pivot irrigation system in Brazil](https://github.com/saraivaufc/instance-segmentation-maskrcnn) using free Landsat images, mask R-CNN \u0026 Keras\n\n- [Building-Detection-MaskRCNN](https://github.com/Mstfakts/Building-Detection-MaskRCNN) -\u003e Building detection from the SpaceNet dataset by using Mask RCNN\n\n- [Oil tank instance segmentation with Mask R-CNN](https://github.com/georgiosouzounis/instance-segmentation-mask-rcnn) with [accompanying article](https://medium.com/@georgios.ouzounis/oil-storage-tank-instance-segmentation-with-mask-r-cnn-77c94433045f) using Keras \u0026 Airbus Oil Storage Detection Dataset on Kaggle\n\n- [Mask_RCNN-for-Caravans](https://github.com/OrdnanceSurvey/Mask_RCNN-for-Caravans) -\u003e detect caravan footprints from OS imagery\n\n- [parking_bays_detectron2](https://github.com/spiyer99/parking_bays_detectron2) -\u003e Detecting parking bays with satellite imagery. Used Detectron2 and synthetic data with Unreal, superior performance to using Mask RCNN\n\n- [Circle_Finder](https://github.com/zinsmatt/Circle_Finder) -\u003e Circular Shapes Detection in Satellite Imagery, 2nd place solution to the Circle Finder Challenge\n\n- [Lawn_maskRCNN](https://github.com/matthewnaples/Lawn_maskRCNN) -\u003e Detecting lawns from satellite images of properties in the Cedar Rapids area using Mask-R-CNN\n\n- [CropMask_RCNN](https://github.com/ecohydro/CropMask_RCNN) -\u003e Segmenting center pivot agriculture to monitor crop water use in drylands with Mask R-CNN and Landsat satellite imagery\n\n- [Mask RCNN for Spacenet Off Nadir Building Detection](https://github.com/ashnair1/Mask-RCNN-for-Off-Nadir-Building-Detection)\n\n- [CATNet](https://github.com/yeliudev/CATNet) -\u003e Learning to Aggregate Multi-Scale Context for Instance Segmentation in Remote Sensing Images\n\n- [Object-Detection-on-Satellite-Images-using-Mask-R-CNN](https://github.com/ThayN15/Object-Detection-on-Satellite-Images-using-Mask-R-CNN) -\u003e detect ships\n\n- [FactSeg](https://github.com/Junjue-Wang/FactSeg) -\u003e Foreground Activation Driven Small Object Semantic Segmentation in Large-Scale Remote Sensing Imagery (TGRS), also see [FarSeg](https://github.com/Z-Zheng/FarSeg) and [FreeNet](https://github.com/Z-Zheng/FreeNet), implementations of research paper\n\n- [aqua_python](https://github.com/tclavelle/aqua_python) -\u003e detecting aquaculture farms using Mask R-CNN\n\n- [RSPrompter](https://github.com/KyanChen/RSPrompter) -\u003e Learning to Prompt for Remote Sensing Instance Segmentation based on Visual Foundation Model\n\n#\n## Object detection\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/object-detection.png\" width=\"600\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eImage showing the suitability of rotated bounding boxes in remote sensing.\u003c/b\u003e\n\u003c/p\u003e\n\nObject detection in remote sensing involves locating and surrounding objects of interest with bounding boxes. Due to the large size of remote sensing images and the fact that objects may only comprise a few pixels, object detection can be challenging in this context. The imbalance between the area of the objects to be detected and the background, combined with the potential for objects to be easily confused with random features in the background, further complicates the task. Object detection generally performs better on larger objects, but becomes increasingly difficult as the objects become smaller and more densely packed. The accuracy of object detection models can also degrade rapidly as image resolution decreases, which is why it is common to use high resolution imagery, such as 30cm RGB, for object detection in remote sensing. A unique characteristic of aerial images is that objects can be oriented in any direction. To effectively extract measurements of the length and width of an object, it can be crucial to use rotated bounding boxes that align with the orientation of the object. This approach enables more accurate and meaningful analysis of the objects within the image. [Image source](https://www.mdpi.com/2072-4292/13/21/4291)\n\n### Object tracking in videos\n\n- [TCTrack](https://github.com/vision4robotics/TCTrack) -\u003e Temporal Contexts for Aerial Tracking\n\n- [CFME](https://github.com/SY-Xuan/CFME) -\u003e Object Tracking in Satellite Videos by Improved Correlation Filters With Motion Estimations\n\n- [TGraM](https://github.com/HeQibin/TGraM) -\u003e Multi-Object Tracking in Satellite Videos with Graph-Based Multi-Task Modeling\n\n- [satellite_video_mod_groundtruth](https://github.com/zhangjunpeng9354/satellite_video_mod_groundtruth) -\u003e groundtruth on satellite video for evaluating moving object detection algorithm\n\n- [Moving-object-detection-DSFNet](https://github.com/ChaoXiao12/Moving-object-detection-DSFNet) -\u003e DSFNet: Dynamic and Static Fusion Network for Moving Object Detection in Satellite Videos\n\n- [HiFT](https://github.com/vision4robotics/HiFT) -\u003e Hierarchical Feature Transformer for Aerial Tracking\n\n### Object detection with rotated bounding boxes\n\nOrinted bounding boxes (OBB) are polygons representing rotated rectangles. For datasets checkout DOTA \u0026 HRSC2016. Start with Yolov8\n\n- [mmrotate](https://github.com/open-mmlab/mmrotate) -\u003e Rotated Object Detection Benchmark, with pretrained models and function for inferencing on very large images\n\n- [OBBDetection](https://github.com/jbwang1997/OBBDetection) -\u003e an oriented object detection library, which is based on MMdetection\n\n- [rotate-yolov3](https://github.com/ming71/rotate-yolov3) -\u003e Rotation object detection implemented with yolov3. Also see [yolov3-polygon](https://github.com/ming71/yolov3-polygon)\n\n- [DRBox](https://github.com/liulei01/DRBox) -\u003e for detection tasks where the objects are orientated arbitrarily, e.g. vehicles, ships and airplanes\n\n- [s2anet](https://github.com/csuhan/s2anet) -\u003e Align Deep Features for Oriented Object Detection\n\n- [CFC-Net](https://github.com/ming71/CFC-Net) -\u003e A Critical Feature Capturing Network for Arbitrary-Oriented Object Detection in Remote Sensing Images\n\n- [ReDet](https://github.com/csuhan/ReDet) -\u003e A Rotation-equivariant Detector for Aerial Object Detection\n\n- [BBAVectors-Oriented-Object-Detection](https://github.com/yijingru/BBAVectors-Oriented-Object-Detection) -\u003e Oriented Object Detection in Aerial Images with Box Boundary-Aware Vectors\n\n- [CSL_RetinaNet_Tensorflow](https://github.com/Thinklab-SJTU/CSL_RetinaNet_Tensorflow) -\u003e Arbitrary-Oriented Object Detection with Circular Smooth Label\n\n- [r3det-on-mmdetection](https://github.com/SJTU-Thinklab-Det/r3det-on-mmdetection) -\u003e R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object\n\n- [R-DFPN_FPN_Tensorflow](https://github.com/yangxue0827/R-DFPN_FPN_Tensorflow) -\u003e Rotation Dense Feature Pyramid Networks (Tensorflow)\n\n- [R2CNN_Faster-RCNN_Tensorflow](https://github.com/DetectionTeamUCAS/R2CNN_Faster-RCNN_Tensorflow) -\u003e Rotational region detection based on Faster-RCNN\n\n- [Rotated-RetinaNet](https://github.com/ming71/Rotated-RetinaNet) -\u003e implemented in pytorch, it supports the following datasets: DOTA, HRSC2016, ICDAR2013, ICDAR2015, UCAS-AOD, NWPU VHR-10, VOC2007\n\n- [OBBDet_Swin](https://github.com/ming71/OBBDet_Swin) -\u003e The sixth place winning solution in 2021 Gaofen Challenge\n\n- [CG-Net](https://github.com/WeiZongqi/CG-Net) -\u003e Learning Calibrated-Guidance for Object Detection in Aerial Images\n\n- [OrientedRepPoints_DOTA](https://github.com/hukaixuan19970627/OrientedRepPoints_DOTA) -\u003e Oriented RepPoints + Swin Transformer/ReResNet\n\n- [yolov5_obb](https://github.com/hukaixuan19970627/yolov5_obb) -\u003e yolov5 + Oriented Object Detection\n\n- [How to Train YOLOv5 OBB](https://blog.roboflow.com/yolov5-for-oriented-object-detection/) -\u003e YOLOv5 OBB tutorial and [YOLOv5 OBB noteboook](https://colab.research.google.com/drive/16nRwsioEYqWFLBF5VpT_NvELeOeupURM#scrollTo=1NZxhXTMWvek)\n\n- [OHDet_Tensorflow](https://github.com/SJTU-Thinklab-Det/OHDet_Tensorflow) -\u003e can be applied to rotation detection and object heading detection\n\n- [Seodore](https://github.com/nijkah/Seodore) -\u003e framework maintaining recent updates of mmdetection\n\n- [Rotation-RetinaNet-PyTorch](https://github.com/HsLOL/Rotation-RetinaNet-PyTorch) -\u003e oriented detector Rotation-RetinaNet implementation on Optical and SAR ship dataset\n\n- [AIDet](https://github.com/jwwangchn/aidet) -\u003e an open source object detection in aerial image toolbox based on MMDetection\n\n- [rotation-yolov5](https://github.com/BossZard/rotation-yolov5) -\u003e rotation detection based on yolov5\n\n- [ShipDetection](https://github.com/lilinhao/ShipDetection) -\u003e Ship Detection in HR Optical Remote Sensing Images via Rotated Bounding Box, based on Faster R-CNN and ORN, uses caffe\n\n- [SLRDet](https://github.com/LUCKMOONLIGHT/SLRDet) -\u003e project based on mmdetection to reimplement RRPN and use the model Faster R-CNN OBB\n\n- [AxisLearning](https://github.com/RSIA-LIESMARS-WHU/AxisLearning) -\u003e Axis Learning for Orientated Objects Detection in Aerial Images\n\n- [Detection_and_Recognition_in_Remote_Sensing_Image](https://github.com/whywhs/Detection_and_Recognition_in_Remote_Sensing_Image) -\u003e This work uses PaNet to realize Detection and Recognition in Remote Sensing Image by MXNet\n\n- [DrBox-v2-tensorflow](https://github.com/ZongxuPan/DrBox-v2-tensorflow) -\u003e tensorflow implementation of DrBox-v2 which is an improved detector with rotatable boxes for target detection in remote sensing images\n\n- [Rotation-EfficientDet-D0](https://github.com/HsLOL/Rotation-EfficientDet-D0) -\u003e A PyTorch Implementation Rotation Detector based EfficientDet Detector, applied to custom rotation vehicle datasets\n\n- [DODet](https://github.com/yanqingyao1994/DODet) -\u003e Dual alignment for oriented object detection, uses DOTA dataset\n\n- [GF-CSL](https://github.com/WangJian981002/GF-CSL) -\u003e Gaussian Focal Loss: Learning Distribution Polarized Angle Prediction for Rotated Object Detection in Aerial Images\n\n- [simplified_rbox_cnn](https://github.com/SIAnalytics/simplified_rbox_cnn) -\u003e RBox-CNN: rotated bounding box based CNN for ship detection in remote sensing image. Uses Tensorflow object detection API\n\n- [Polar-Encodings](https://github.com/flyingshan/Learning-Polar-Encodings-For-Arbitrary-Oriented-Ship-Detection-In-SAR-Images) -\u003e Learning Polar Encodings for Arbitrary-Oriented Ship Detection in SAR Images\n\n- [R-CenterNet](https://github.com/ZeroE04/R-CenterNet) -\u003e detector for rotated-object based on CenterNet\n\n- [piou](https://github.com/clobotics/piou) -\u003e Orientated Object Detection; IoU Loss, applied to DOTA dataset\n\n- [DAFNe](https://github.com/steven-lang/DAFNe) -\u003e A One-Stage Anchor-Free Approach for Oriented Object Detection\n\n- [AProNet](https://github.com/geovsion/AProNet) -\u003e Detecting objects with precise orientation from aerial images. Applied to datasets DOTA and HRSC2016\n\n- [UCAS-AOD-benchmark](https://github.com/ming71/UCAS-AOD-benchmark) -\u003e A benchmark of UCAS-AOD dataset\n\n- [RotateObjectDetection](https://github.com/XinzeLee/RotateObjectDetection) -\u003e based on Ultralytics/yolov5, with adjustments to enable rotate prediction boxes. Also see [PolygonObjectDetection](https://github.com/XinzeLee/PolygonObjectDetection)\n\n- [AD-Toolbox](https://github.com/liuyanyi/AD-Toolbox) -\u003e Aerial Detection Toolbox based on MMDetection and MMRotate, with support for more datasets\n\n- [GGHL](https://github.com/Shank2358/GGHL) -\u003e A General Gaussian Heatmap Label Assignment for Arbitrary-Oriented Object Detection\n\n- [NPMMR-Det](https://github.com/Shank2358/NPMMR-Det) -\u003e A Novel Nonlocal-Aware Pyramid and Multiscale Multitask Refinement Detector for Object Detection in Remote Sensing Images\n\n- [AOPG](https://github.com/jbwang1997/AOPG) -\u003e Anchor-Free Oriented Proposal Generator for Object Detection\n\n- [SE2-Det](https://github.com/Virusxxxxxxx/SE2-Det) -\u003e Semantic-Edge-Supervised Single-Stage Detector for Oriented Object Detection in Remote Sensing Imagery\n\n- [OrientedRepPoints](https://github.com/LiWentomng/OrientedRepPoints) -\u003e Oriented RepPoints for Aerial Object Detection\n\n- [TS-Conv](https://github.com/Shank2358/TS-Conv) -\u003e Task-wise Sampling Convolutions for Arbitrary-Oriented Object Detection in Aerial Images\n\n- [FCOSR](https://github.com/lzh420202/FCOSR) -\u003e A Simple Anchor-free Rotated Detector for Aerial Object Detection. This implement is modified from mmdetection. See also [TensorRT_Inference](https://github.com/lzh420202/TensorRT_Inference)\n\n- [OBB_Detection](https://github.com/HsLOL/OBB_Detection) -\u003e Finalist's solution in the track of Oriented Object Detection in Remote Sensing Images, 2022 Guangdong-Hong Kong-Macao Greater Bay Area International Algorithm Competition\n\n- [sam-mmrotate](https://github.com/Li-Qingyun/sam-mmrotate) -\u003e SAM (Segment Anything Model) for generating rotated bounding boxes with MMRotate, which is a comparison method of H2RBox-v2\n\n- [mmrotate-dcfl](https://github.com/Chasel-Tsui/mmrotate-dcfl) -\u003e Dynamic Coarse-to-Fine Learning for Oriented Tiny Object Detection\n\n- [h2rbox-mmrotate](https://github.com/yangxue0827/h2rbox-mmrotate) -\u003e Horizontal Box Annotation is All You Need for Oriented Object Detection\n\n- [Spatial-Transform-Decoupling](https://github.com/yuhongtian17/Spatial-Transform-Decoupling) -\u003e Spatial Transform Decoupling for Oriented Object Detection\n\n- [ARS-DETR](https://github.com/httle/ARS-DETR) -\u003e Aspect Ratio Sensitive Oriented Object Detection with Transformer\n\n- [CFINet](https://github.com/shaunyuan22/CFINet) -\u003e Small Object Detection via Coarse-to-fine Proposal Generation and Imitation Learning. Introduces [SODA-A dataset](https://shaunyuan22.github.io/SODA/)\n\n### Object detection enhanced by super resolution\n\n- [Super-Resolution and Object Detection](https://medium.com/the-downlinq/super-resolution-and-object-detection-a-love-story-part-4-8ad971eef81e) -\u003e Super-resolution is a relatively inexpensive enhancement that can improve object detection performance\n\n- [EESRGAN](https://github.com/Jakaria08/EESRGAN) -\u003e Small-Object Detection in Remote Sensing Images with End-to-End Edge-Enhanced GAN and Object Detector Network\n\n- [Mid-Low Resolution Remote Sensing Ship Detection Using Super-Resolved Feature Representation](https://www.preprints.org/manuscript/202108.0337/v1)\n\n- [EESRGAN](https://github.com/divyam96/EESRGAN) -\u003e Small-Object Detection in Remote Sensing Images with End-to-End Edge-Enhanced GAN and Object Detector Network. Applied to COWC \u0026 [OGST](https://data.mendeley.com/datasets/bkxj8z84m9/3) datasets\n\n- [FBNet](https://github.com/wdzhao123/FBNet) -\u003e Feature Balance for Fine-Grained Object Classification in Aerial Images\n\n- [SuperYOLO](https://github.com/icey-zhang/SuperYOLO) -\u003e SuperYOLO: Super Resolution Assisted Object Detection in Multimodal Remote Sensing Imagery\n\n### Salient object detection\n\nDetecting the most noticeable or important object in a scene\n\n- [ACCoNet](https://github.com/MathLee/ACCoNet) -\u003e Adjacent Context Coordination Network for Salient Object Detection in Optical Remote Sensing Images\n\n- [MCCNet](https://github.com/MathLee/MCCNet) -\u003e Multi-Content Complementation Network for Salient Object Detection in Optical Remote Sensing Images\n\n- [CorrNet](https://github.com/MathLee/CorrNet) -\u003e Lightweight Salient Object Detection in Optical Remote Sensing Images via Feature Correlation\n\n- [Reading list for deep learning based Salient Object Detection in Optical Remote Sensing Images](https://github.com/MathLee/ORSI-SOD_Summary)\n\n- [ORSSD-dataset](https://github.com/rmcong/ORSSD-dataset) -\u003e salient object detection dataset\n\n- [EORSSD-dataset](https://github.com/rmcong/EORSSD-dataset) -\u003e Extended Optical Remote Sensing Saliency Detection (EORSSD) Dataset\n\n- [DAFNet_TIP20](https://github.com/rmcong/DAFNet_TIP20) -\u003e Dense Attention Fluid Network for Salient Object Detection in Optical Remote Sensing Images\n\n- [EMFINet](https://github.com/Kunye-Shen/EMFINet) -\u003e Edge-Aware Multiscale Feature Integration Network for Salient Object Detection in Optical Remote Sensing Images\n\n- [ERPNet](https://github.com/zxforchid/ERPNet) -\u003e Edge-guided Recurrent Positioning Network for Salient Object Detection in Optical Remote Sensing Images\n\n- [FSMINet](https://github.com/zxforchid/FSMINet) -\u003e Fully Squeezed Multi-Scale Inference Network for Fast and Accurate Saliency Detection in Optical Remote Sensing Images\n\n- [AGNet](https://github.com/NuaaYH/AGNet) -\u003e AGNet: Attention Guided Network for Salient Object Detection in Optical Remote Sensing Images\n\n- [MSCNet](https://github.com/NuaaYH/MSCNet) -\u003e A lightweight multi-scale context network for salient object detection in optical remote sensing images\n\n- [GPnet](https://github.com/liuyu1002/GPnet) -\u003e Global Perception Network for Salient Object Detection in Remote Sensing Images\n\n- [SeaNet](https://github.com/MathLee/SeaNet) -\u003e Lightweight Salient Object Detection in Optical Remote Sensing Images via Semantic Matching and Edge Alignment\n\n- [GeleNet](https://github.com/MathLee/GeleNet) -\u003e Salient Object Detection in Optical Remote Sensing Images Driven by Transformer\n\n### Object detection - Buildings, rooftops \u0026 solar panels\n\n- [satellite_image_tinhouse_detector](https://github.com/yasserius/satellite_image_tinhouse_detector) -\u003e Detection of tin houses from satellite/aerial images using the Tensorflow Object Detection API\n\n- [Machine Learning For Rooftop Detection and Solar Panel Installment](https://omdena.com/blog/machine-learning-rooftops/) discusses tiling large images and generating annotations from OSM data. Features of the roofs were calculated using a combination of contour detection and classification. [Follow up article using semantic segmentation](https://omdena.com/blog/rooftops-classification/)\n\n- [Building Extraction with YOLT2 and SpaceNet Data](https://medium.com/the-downlinq/building-extraction-with-yolt2-and-spacenet-data-a926f9ffac4f)\n\n- [XBD-hurricanes](https://github.com/dbuscombe-usgs/XBD-hurricanes) -\u003e Models for building (and building damage) detection in high-resolution (\u003c1m) satellite and aerial imagery using a modified RetinaNet model\n\n- [ssd-spacenet](https://github.com/aurotripathy/ssd-spacenet) -\u003e Detect buildings in the Spacenet dataset using Single Shot MultiBox Detector (SSD)\n\n- [3DBuildingInfoMap](https://github.com/LllC-mmd/3DBuildingInfoMap) -\u003e simultaneous extraction of building height and footprint from Sentinel imagery using ResNet\n\n- [DeepSolaris](https://github.com/thinkpractice/DeepSolaris) -\u003e a EuroStat project to detect solar panels in aerial images, further material [here](https://github.com/FHNW-IVGI/workshop_geopython2019/tree/master/Ex.02_SolarPanels)\n\n- [ML_ObjectDetection_CAFO](https://github.com/Qberto/ML_ObjectDetection_CAFO) -\u003e Detect Concentrated Animal Feeding Operations (CAFO) in Satellite Imagery\n\n- [Multi-level-Building-Detection-Framework](https://github.com/luoxiaoliaolan/Multi-level-Building-Detection-Framework) -\u003e Multilevel Building Detection Framework in Remote Sensing Images Based on Convolutional Neural Networks\n\n- [Automatic Damage Annotation on Post-Hurricane Satellite Imagery](https://dds-lab.github.io/disaster-damage-detection/) -\u003e detect damaged buildings using tensorflow object detection API. With repos [here](https://github.com/DDS-Lab/disaster-image-processing) and [here](https://github.com/annieyan/PreprocessSatelliteImagery-ObjectDetection)\n\n- [mappingchallenge](https://github.com/krishanr/mappingchallenge) -\u003e YOLOv5 applied to the AICrowd Mapping Challenge dataset\n\n### Object detection - Ships, boats, vessels \u0026 wake\n\n- [Airbus Ship Detection Challenge](https://www.kaggle.com/c/airbus-ship-detection) -\u003e using oriented bounding boxes. Read [Detecting ships in satellite imagery: five years later…](https://medium.com/artificialis/detecting-ships-in-satellite-imagery-five-years-later-28df2e83f987)\n\n- [kaggle-ships-in-Google-Earth-yolov8](https://github.com/robmarkcole/kaggle-ships-in-satellite-imagery-with-YOLOv8) -\u003e Applying YOLOv8 to Kaggle Ships in Google Earth dataset\n\n- [How hard is it for an AI to detect ships on satellite images?](https://medium.com/earthcube-stories/how-hard-it-is-for-an-ai-to-detect-ships-on-satellite-images-7265e34aadf0)\n\n- [Object Detection in Satellite Imagery, a Low Overhead Approach](https://medium.com/the-downlinq/object-detection-in-satellite-imagery-a-low-overhead-approach-part-i-cbd96154a1b7)\n\n- [Detecting Ships in Satellite Imagery](https://medium.com/dataseries/detecting-ships-in-satellite-imagery-7f0ca04e7964) using the Planet dataset and Keras\n\n- [SARfish](https://github.com/MJCruickshank/SARfish) -\u003e Ship detection in Sentinel 1 Synthetic Aperture Radar (SAR) imagery\n\n- [Arbitrary-Oriented Ship Detection through Center-Head Point Extraction](https://github.com/JinleiMa/ASD)\n\n- [ship_detection](https://github.com/rugg2/ship_detection) -\u003e using an interesting combination of CNN classifier, Class Activation Mapping (CAM) \u0026 UNET segmentation\n\n- [Building a complete Ship detection algorithm using YOLOv3 and Planet satellite images](https://medium.com/intel-software-innovators/ship-detection-in-satellite-images-from-scratch-849ccfcc3072) -\u003e covers finding and annotating data (using LabelMe), preprocessing large images into chips, and training Yolov3. [Repo](https://github.com/amanbasu/ship-detection)\n\n- [Ship-detection-in-satellite-images](https://github.com/zmf0507/Ship-detection-in-satellite-images) -\u003e experiments with  UNET, YOLO, Mask R-CNN, SSD, Faster R-CNN, RETINA-NET\n\n- [Ship-Detection-from-Satellite-Images-using-YOLOV4](https://github.com/debasis-dotcom/Ship-Detection-from-Satellite-Images-using-YOLOV4) -\u003e uses Kaggle Airbus Ship Detection dataset\n\n- [shipsnet-detector](https://github.com/rhammell/shipsnet-detector) -\u003e Detect container ships in Planet imagery using machine learning\n\n- [Mask R-CNN for Ship Detection \u0026 Segmentation](https://medium.com/@gabogarza/mask-r-cnn-for-ship-detection-segmentation-a1108b5a083) blog post with [repo](https://github.com/gabrielgarza/Mask_RCNN)\n\n- [contrastive_SSL_ship_detection](https://github.com/alina2204/contrastive_SSL_ship_detection) -\u003e Contrastive self supervised learning for ship detection in Sentinel 2 images\n\n- [Boat detection with multi-region-growing method in satellite images](https://medium.com/@ipmach/boat-detection-with-multi-region-growing-method-in-satellite-images-3339a6c29a8c)\n\n- [small-boat-detector](https://github.com/swricci/small-boat-detector) -\u003e Trained yolo v3 model weights and configuration file to detect small boats in satellite imagery\n\n- [Satellite-Imagery-Datasets-Containing-Ships](https://github.com/JasonManesis/Satellite-Imagery-Datasets-Containing-Ships) -\u003e A list of optical and radar satellite datasets for ship detection, classification, semantic segmentation and instance segmentation tasks\n\n- [vessel-detection-sentinels](https://github.com/allenai/vessel-detection-sentinels) -\u003e Sentinel-1 and Sentinel-2 Vessel Detection\n\n- [Ship-Detection](https://github.com/gouravbarkle/Ship-Detection) -\u003e CNN approach for ship detection in the ocean using a satellite image\n\n- [vesselTracker](https://github.com/carlossantamarizq/vesselTracker) -\u003e Project based on reduced model of Yolov5 architecture using Pytorch. Custom dataset based on SAR imagery provided by Sentinel-1 through Earth Engine API\n\n- [marine-debris-ml-model](https://github.com/danieltyukov/marine-debris-ml-model) -\u003e Marine Debris Detection using tensorflow object detection API\n\n- [SDGH-Net](https://github.com/WangZhenqing-RS/SDGH-Net-Ship-Detection-in-Optical-Remote-Sensing-Images-Based-on-Gaussian-Heatmap-Regression) -\u003e Ship Detection in Optical Remote Sensing Images Based on Gaussian Heatmap Regression\n\n- [LR-TSDet](https://github.com/Lausen-Ng/LR-TSDet) -\u003e LR-TSDet: Towards Tiny Ship Detection in Low-Resolution Remote Sensing Images\n\n- [FGSCR-42](https://github.com/DYH666/FGSCR-42) -\u003e A public Dataset for Fine-Grained Ship Classification in Remote sensing images\n\n- [ShipDetection](https://github.com/lilinhao/ShipDetection) -\u003e Ship Detection in HR Optical Remote Sensing Images via Rotated Bounding Box, based on Faster R-CNN and ORN, uses caffe\n\n- [WakeNet](https://github.com/Lilytopia/WakeNet) -\u003e Rethinking Automatic Ship Wake Detection: State-of-the-Art CNN-based Wake Detection via Optical Images\n\n- [Histogram of Oriented Gradients (HOG) Boat Heading Classification](https://medium.com/the-downlinq/histogram-of-oriented-gradients-hog-heading-classification-a92d1cf5b3cc)\n\n- [Object Detection in Satellite Imagery, a Low Overhead Approach](https://medium.com/the-downlinq/object-detection-in-satellite-imagery-a-low-overhead-approach-part-i-cbd96154a1b7) -\u003e Medium article which demonstrates how to combine Canny edge detector pre-filters with HOG feature descriptors, random forest classifiers, and sliding windows to perform ship detection\n\n- [simplified_rbox_cnn](https://github.com/SIAnalytics/simplified_rbox_cnn) -\u003e RBox-CNN: rotated bounding box based CNN for ship detection in remote sensing image. Uses Tensorflow object detection API\n\n- [Ship-Detection-based-on-YOLOv3-and-KV260](https://github.com/xlsjdjdk/Ship-Detection-based-on-YOLOv3-and-KV260) -\u003e entry project of the Xilinx Adaptive Computing Challenge 2021. It uses YOLOv3 for ship target detection in optical remote sensing images, and deploys DPU on the KV260 platform to achieve hardware acceleration\n\n- [LEVIR-Ship](https://github.com/WindVChen/LEVIR-Ship) -\u003e a dataset for tiny ship detection under medium-resolution remote sensing images\n\n- [Push-and-Pull-Network](https://github.com/WindVChen/Push-and-Pull-Network) -\u003e Contrastive Learning for Fine-grained Ship Classification in Remote Sensing Images\n\n- [DRENet](https://github.com/WindVChen/DRENet) -\u003e A Degraded Reconstruction Enhancement-Based Method for Tiny Ship Detection in Remote Sensing Images With a New Large-Scale Dataset\n\n- [xView3-The-First-Place-Solution](https://github.com/BloodAxe/xView3-The-First-Place-Solution) -\u003e A winning solution for [xView 3](https://iuu.xview.us/) challenge (Vessel detection, classification and length estimation on Sentinetl-1 images). Contains trained models, inference pipeline and training code \u0026 configs to reproduce the results.\n\n- [vessel-detection-viirs](https://github.com/allenai/vessel-detection-viirs) -\u003e Model and service code for streaming vessel detections from VIIRS satellite imagery\n\n- [wakemodel_llmassist](https://github.com/pbeukema/wakemodel_llmassist) -\u003e wake detection in Sentinel-2, uses an EfficientNet-B0 architecture adapted for keypoint detection\n\n- [ORFENet](https://github.com/dyl96/ORFENet) -\u003e Tiny Object Detection in Remote Sensing Images Based on Object Reconstruction and Multiple Receptive Field Adaptive Feature Enhancement. Uses LEVIR-Ship \u0026 AI-TODv2 datasets\n\n- [mayrajeo S2 ship-detection](https://github.com/mayrajeo/ship-detection) -\u003e Detecting marine vessels from Sentinel-2 imagery with YOLOv8\n\n- [CHPDet](https://github.com/zf020114/CHPDet) -\u003e PyTorch implementation of \"Arbitrary-Oriented Ship Detection through Center-Head Point Extraction\"\n\n- [VDS2Raw](https://github.com/sirbastiano/VDS2Raw) -\u003e VFNet with ResNet-18 for Vessel Detection in S-2 Raw Imagery\n\n- [Global Fishing Capacity - Vessel Detection Model](https://github.com/allenai/global_fishing_capacity_detector) -\u003e from Allen.ai and using Maxar imagery\n\n### Object detection - Cars, vehicles \u0026 trains\n\n- [Detection of parkinglots and driveways with retinanet](https://github.com/spiyer99/retinanet)\n\n- [pytorch-vedai](https://github.com/MichelHalmes/pytorch-vedai) -\u003e object detection on the VEDAI dataset: Vehicle Detection in Aerial Imagery\n\n- [Truck Detection with Sentinel-2 during COVID-19 crisis](https://github.com/hfisser/Truck_Detection_Sentinel2_COVID19) -\u003e moving objects in Sentinel-2 data causes a specific reflectance relationship in the RGB, which looks like a rainbow, and serves as a marker for trucks. Improve accuracy by only analysing roads. Not using object detection but relevant. Also see [S2TD](https://github.com/hfisser/S2TD)\n\n- [cowc_car_counting](https://github.com/motokimura/cowc_car_counting) -\u003e car counting on the [Cars Overhead With Context (COWC) dataset](https://gdo152.llnl.gov/cowc/). Not sctictly object detection but a CNN to predict the car count in a tile\n\n- [CarCounting](https://github.com/JacksonPeoples/CarCounting) -\u003e using Yolov3 \u0026 COWC dataset\n\n- [Traffic density estimation as a regression problem instead of object detection](https://omdena.com/blog/ai-road-safety/)\n\n- [Rotation-EfficientDet-D0](https://github.com/HsLOL/Rotation-EfficientDet-D0) -\u003e PyTorch implementation of Rotated EfficientDet, applied to a custom rotation vehicle dataset (car counting)\n\n- [RSVC2021-Dataset](https://github.com/YinongGuo/RSVC2021-Dataset) -\u003e A dataset for Vehicle Counting in Remote Sensing images, created from the DOTA \u0026 ITCVD\n\n- [Car Localization and Counting with Overhead Imagery, an Interactive Exploration](https://medium.com/the-downlinq/car-localization-and-counting-with-overhead-imagery-an-interactive-exploration-9d5a029a596b) -\u003e Medium article by Adam Van Etten\n\n- [Vehicle-Counting-in-Very-Low-Resolution-Aerial-Images](https://github.com/hbsszq/Vehicle-Counting-in-Very-Low-Resolution-Aerial-Images) -\u003e Vehicle Counting in Very Low-Resolution Aerial Images via Cross-Resolution Spatial Consistency and Intraresolution Time Continuity\n\n- [Vehicle Detection blog post](https://www.silvispace.xyz/posts/vehicle-post/) by Grant Pearse: detecting vehicles across New Zealand without collecting local training data\n\n- [detecting-trucks](https://github.com/datasciencecampus/detecting-trucks) -\u003e detecting large vehicles in Sentinel-2\n\n### Object detection - Planes \u0026 aircraft\n- [FlightScope_Bench](https://github.com/toelt-llc/FlightScope_Bench) -\u003e A Deep Comprehensive Assessment of Aircraft Detection Algorithms in Satellite Imagery, including Faster RCNN, DETR, SSD, RTMdet, RetinaNet, CenterNet, YOLOv5, and YOLOv8\n\n- [Faster RCNN to detect airplanes](https://github.com/ShubhankarRawat/Airplane-Detection-for-Satellites)\n\n- [yoltv4](https://github.com/avanetten/yoltv4) includes examples on the [RarePlanes dataset](https://registry.opendata.aws/rareplanes/)\n\n- [aircraft-detection](https://github.com/hakeemtfrank/aircraft-detection) -\u003e experiments to test the performance of a Gaussian process (GP) classifier with various kernels on the UC Merced land use land cover (LULC) dataset\n\n- [aircraft-detection-from-satellite-images-yolov3](https://github.com/emrekrtorun/aircraft-detection-from-satellite-images-yolov3) -\u003e trained on kaggle cgi-planes-in-satellite-imagery-w-bboxes dataset\n\n- [HRPlanesv2-Data-Set](https://github.com/dilsadunsal/HRPlanesv2-Data-Set) -\u003e YOLOv4 and YOLOv5 weights trained on the HRPlanesv2 dataset\n\n- [Deep-Learning-for-Aircraft-Recognition](https://github.com/Shayan-Bravo/Deep-Learning-for-Aircraft-Recognition) -\u003e A CNN model trained to classify and identify various military aircraft through satellite imagery\n\n- [FRCNN-for-Aircraft-Detection](https://github.com/Huatsing-Lau/FRCNN-for-Aircraft-Detection)\n\n- [ergo-planes-detector](https://github.com/evilsocket/ergo-planes-detector) -\u003e An ergo based project that relies on a convolutional neural network to detect airplanes from satellite imagery, uses the PlanesNet dataset\n\n- [pytorch-remote-sensing](https://github.com/miko7879/pytorch-remote-sensing) -\u003e Aircraft detection using the 'Airbus Aircraft Detection' dataset and Faster-RCNN with ResNet-50 backbone using pytorch\n\n- [FasterRCNN_ObjectDetection](https://github.com/UKMIITB/FasterRCNN_ObjectDetection) -\u003e faster RCNN model for aircraft detection and localisation in satellite images and creating a webpage with live server for public usage\n\n- [HRPlanes](https://github.com/TolgaBkm/HRPlanes) -\u003e weights of YOLOv4 and Faster R-CNN networks trained with HRPlanes dataset\n\n- [aerial-detection](https://github.com/alexbakr/aerial-detection) -\u003e uses Yolov5 \u0026 Icevision\n\n- [How to choose a deep learning architecture to detect aircrafts in satellite imagery?](https://medium.com/artificialis/how-to-choose-a-deep-learning-model-to-detect-aircrafts-in-satellite-imagery-cd7d106e76ad)\n\n- [rareplanes-yolov5](https://github.com/jeffaudi/rareplanes-yolov5) -\u003e using YOLOv5 and the RarePlanes dataset to detect and classify sub-characteristics of aircraft, with [article](https://medium.com/artificialis/detecting-aircrafts-on-airbus-pleiades-imagery-with-yolov5-5f3d464b75ad)\n\n- [OnlyPlanes](https://github.com/naivelogic/OnlyPlanes) -\u003e Incrementally Tuning Synthetic Training Datasets for Satellite Object Detection\n\n- [Understanding the RarePlanes Dataset and Building an Aircraft Detection Model](https://encord.com/blog/rareplane-dataset-aircraft-detection-model/) -\u003e blog post\n\n### Object detection - Infrastructure \u0026 utilities\n\n- [wind-turbine-detector](https://github.com/lbborkowski/wind-turbine-detector) -\u003e Wind Turbine Object Detection from Aerial Imagery Using TensorFlow Object Detection API\n\n- [Water Tanks and Swimming Pools Detection](https://github.com/EduardoFernandes1410/PATREO-Dengue) -\u003e uses Faster R-CNN\n\n- [PCAN](https://www.mdpi.com/2072-4292/13/7/1243) -\u003e Part-Based Context Attention Network for Thermal Power Plant Detection in Remote Sensing Imagery, with [dataset](https://github.com/wenxinYin/AIR-TPPDD)\n\n- [WindTurbineDetection](https://github.com/nvriese1/WindTurbineDetection) -\u003e Implementation of transfer learning approach using the YOLOv7 framework to detect and rapidly quantify wind turbines in raw LANDSAT and NAIP satellite imagery\n\n- [Arctic-Infrastructure-Detection-Paper](https://github.com/eliasm56/Arctic-Infrastructure-Detection-Paper) -\u003e Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery [paper](https://www.mdpi.com/2072-4292/14/11/2719)\n\n### Object detection - Oil storage tank detection\n\nOil is stored in tanks at many points between extraction and sale, and the volume of oil in storage is an important economic indicator.\n\n- [A Beginner’s Guide To Calculating Oil Storage Tank Occupancy With Help Of Satellite Imagery](https://medium.com/planet-stories/a-beginners-guide-to-calculating-oil-storage-tank-occupancy-with-help-of-satellite-imagery-e8f387200178)\n\n- [Oil-Tank-Volume-Estimation](https://github.com/kheyer/Oil-Tank-Volume-Estimation) -\u003e combines object detection and classical computer vision\n\n- [Oil tank instance segmentation with Mask R-CNN](https://github.com/georgiosouzounis/instance-segmentation-mask-rcnn) with [accompanying article](https://medium.com/@georgios.ouzounis/oil-storage-tank-instance-segmentation-with-mask-r-cnn-77c94433045f) using Keras \u0026 Airbus Oil Storage Detection Dataset on Kaggle\n\n- [SubpixelCircleDetection](https://github.com/anttad/SubpixelCircleDetection) -\u003e CIRCULAR-SHAPED OBJECT DETECTION IN LOW RESOLUTION SATELLITE IMAGES\n\n- [oil_storage-detector](https://github.com/TheodorEmanuelsson/oil_storage-detector) -\u003e using yolov5 and the Airbus Oil Storage Detection dataset\n\n- [oil_well_detector](https://github.com/dzubke/oil_well_detector) -\u003e detect oil wells in the Bakken oil field based on satellite imagery\n\n- [Oil Storage Detection on Airbus Imagery with YOLOX](https://medium.com/artificialis/oil-storage-detection-on-airbus-imagery-with-yolox-9e38eb6f7e62) -\u003e uses the Kaggle Airbus Oil Storage Detection dataset\n\n- [AContrarioTankDetection](https://github.com/anttad/AContrarioTankDetection) -\u003e Oil Tank Detection in Satellite Images via a Contrario Clustering\n\n- [Fast-Large-Image-Object-Detection-yolov7](https://github.com/shah0nawaz/Fast-Large-Image-Object-Detection-yolov7) -\u003e The oil yolov7 model is trained on oil storage tanks (OST) dataset\n\n- [Oiltank-Capacity-Detection](https://github.com/GeNiaaz/Oiltank-Capacity-Detection) -\u003e Analyse storage tanks around the world and identify the external floating roof tanks.\n\n### Object detection - Animals\n\nA variety of techniques can be used to count animals, including object detection and instance segmentation. For convenience they are all listed here:\n\n- [cownter_strike](https://github.com/IssamLaradji/cownter_strike) -\u003e counting cows, located with point-annotations, two models: CSRNet (a density-based method) \u0026 LCFCN (a detection-based method)\n\n- [elephant_detection](https://github.com/akharina/elephant_detection) -\u003e Using Keras-Retinanet to detect elephants from aerial images\n\n- [CNN-Mosquito-Detection](https://github.com/sriramelango/CNN-Mosquito-Detection) -\u003e determining the locations of potentially dangerous breeding grounds, compared YOLOv4, YOLOR \u0026 YOLOv5\n\n- [Borowicz_etal_Spacewhale](https://github.com/lynch-lab/Borowicz_etal_Spacewhale) -\u003e locate whales using ResNet\n\n- [walrus-detection-and-count](https://github.com/sweetlhare/walrus-detection-and-count) -\u003e uses Mask R-CNN instance segmentation\n\n- [MarineMammalsDetection](https://github.com/Pangoraw/MarineMammalsDetection) -\u003e Weakly Supervised Detection of Marine Animals in High Resolution Aerial Images\n\n- [Audubon_F21](https://github.com/RiceD2KLab/Audubon_F21) -\u003e  Deep object detection for waterbird monitoring using aerial imagery\n\n### Object detection - Miscellaneous\n\n- [Object detection on Satellite Imagery using RetinaNet](https://medium.com/@ije_good/object-detection-on-satellite-imagery-using-retinanet-part-1-training-e589975afbd5) -\u003e using the Kaggle Swimming Pool and Car Detection dataset\n\n- [Tackling the Small Object Problem in Object Detection](https://blog.roboflow.com/tackling-the-small-object-problem-in-object-detection)\n\n- [Object Detection and Image Segmentation with Deep Learning on Earth Observation Data: A Review](https://www.mdpi.com/2072-4292/12/10/1667)\n\n- [awesome-aerial-object-detection bu murari023](https://github.com/murari023/awesome-aerial-object-detection), [another by visionxiang](https://github.com/visionxiang/awesome-object-detection-in-aerial-images) and [awesome-tiny-object-detection](https://github.com/kuanhungchen/awesome-tiny-object-detection) list many relevant papers\n\n- [Object Detection Accuracy as a Function of Image Resolution](https://medium.com/the-downlinq/the-satellite-utility-manifold-object-detection-accuracy-as-a-function-of-image-resolution-ebb982310e8c) -\u003e Medium article using COWC dataset, performance rapidly degrades below 30cm imagery\n\n- [Satellite Imagery Multiscale Rapid Detection with Windowed Networks (SIMRDWN)](https://github.com/avanetten/simrdwn) -\u003e combines some of the leading object detection algorithms into a unified framework designed to detect objects both large and small in overhead imagery. Train models and test on arbitrary image sizes with YOLO (versions 2 and 3), Faster R-CNN, SSD, or R-FCN.\n\n- [YOLTv4](https://github.com/avanetten/yoltv4) -\u003e YOLTv4 is designed to detect objects in aerial or satellite imagery in arbitrarily large images that far exceed the ~600×600 pixel size typically ingested by deep learning object detection frameworks\n\n- [Tensorflow Benchmarks for Object Detection in Aerial Images](https://github.com/yangxue0827/RotationDetection)\n\n- [Pytorch Benchmarks for Object Detection in Aerial Images](https://github.com/dingjiansw101/AerialDetection)\n\n- [ASPDNet](https://github.com/liuqingjie/ASPDNet) -\u003e Counting dense objects in remote sensing images\n\n- [xview-yolov3](https://github.com/ultralytics/xview-yolov3) -\u003e xView 2018 Object Detection Challenge: YOLOv3 Training and Inference\n\n- [Faster RCNN for xView satellite data challenge](https://github.com/samirsen/small-object-detection)\n\n- [How to detect small objects in (very) large images](https://blog.ml6.eu/how-to-detect-small-objects-in-very-large-images-70234bab0f98) -\u003e A practical guide to using Slicing-Aided Hyper Inference (SAHI) for performing inference on the DOTAv1.0 object detection dataset using the mmdetection framework\n\n- [Object Detection Satellite Imagery Multi-vehicles Dataset (SIMD)](https://github.com/asimniazi63/Object-Detection-on-Satellite-Images) -\u003e RetinaNet,Yolov3 and Faster RCNN for multi object detection on satellite images dataset\n\n- [SNIPER/AutoFocus](https://github.com/mahyarnajibi/SNIPER) -\u003e an efficient multi-scale object detection training/inference algorithm\n\n- [marine_debris_ML](https://github.com/NASA-IMPACT/marine_debris_ML) -\u003e Marine debris detection, uses 3-meter imagery product called Planetscope with bands in the red, green, blue, and near-infrared. Uses Tensorflow Object Detection API with pre-trained resnet 101\n\n- [Electric-Pylon-Detection-in-RSI](https://github.com/qsjxyz/Electric-Pylon-Detection-in-RSI) -\u003e a dataset which contains 1500 remote sensing images of electric pylons used to train ten deep learning models\n\n- [IS-Count](https://github.com/sustainlab-group/IS-Count) -\u003e IS-Count is a sampling-based and learnable method for estimating the total object count in a region\n\n- [Clustered-Object-Detection-in-Aerial-Image](https://github.com/fyangneil/Clustered-Object-Detection-in-Aerial-Image)\n\n- [yolov5s_for_satellite_imagery](https://github.com/KevinMuyaoGuo/yolov5s_for_satellite_imagery) -\u003e yolov5s applied to the DOTA dataset\n\n- [RetinaNet-PyTorch](https://github.com/HsLOL/RetinaNet-PyTorch) -\u003e RetinaNet implementation on remote sensing ship dataset (SSDD)\n\n- [Detecting-Cyclone-Centers-Custom-YOLOv3](https://github.com/ShubhayanS/Detecting-Cyclone-Centers-Custom-YOLOv3) -\u003e tropical cyclones (TCs) are intense warm-corded cyclonic vortices, developed from low-pressure systems over the tropical oceans and driven by complex air-sea interaction\n\n- [Object-Detection-YoloV3-RetinaNet-FasterRCNN](https://github.com/bostankhan6/Object-Detection-YoloV3-RetinaNet-FasterRCNN) -\u003e trained on a private datset\n\n- [Google-earth-Object-Recognition](https://github.com/InnovAIco/Google-earth-Object-Recognition) -\u003e Code for training and evaluating on Dior Dataset (Google Earth Images) using RetinaNet and YOLOV5\n\n- [HIECTOR: Hierarchical object detector at scale](https://medium.com/sentinel-hub/hiector-hierarchical-object-detector-at-scale-5a61753b51a3) -\u003e HIECTOR facilitates multiple satellite data collections of increasingly detailed spatial resolution for a cost-efficient and accurate object detection over large areas. [Code](https://github.com/sentinel-hub/hiector)\n\n- [Detection of Multiclass Objects in Optical Remote Sensing Images](https://github.com/WenchaoliuMUC/Detection-of-Multiclass-Objects-in-Optical-Remote-Sensing-Images) -\u003e Detection of Multiclass Objects in Optical Remote Sensing Images\n\n- [SB-MSN](https://github.com/weihancug/Sampling-Balance_Multi-stage_Network) -\u003e Improving Training Instance Quality in Aerial Image Object Detection With a Sampling-Balance-Based Multistage Network\n\n- [yoltv5](https://github.com/avanetten/yoltv5) -\u003e detects objects in arbitrarily large aerial or satellite images that far exceed the ~600×600 pixel size typically ingested by deep learning object detection frameworks. Uses YOLOv5 \u0026 pytorch\n\n- [AIR](https://github.com/Accenture/AIR) -\u003e A deep learning object detector framework written in Python for supporting Land Search and Rescue Missions\n\n- [dior_detect](https://github.com/hm-better/dior_detect) -\u003e benchmarks for object detection on DIOR dataset\n\n- [Panchromatic to Multispectral: Object Detection Performance as a Function of Imaging Bands](https://medium.com/the-downlinq/panchromatic-to-multispectral-object-detection-performance-as-a-function-of-imaging-bands-51ecaaa3dc56) -\u003e Medium article, concludes that more bands are not always beneficial, but likely varies by use case\n\n- [OPLD-Pytorch](https://github.com/yf19970118/OPLD-Pytorch) -\u003e Learning Point-Guided Localization for Detection in Remote Sensing Images\n\n- [F3Net](https://github.com/yxhnjust/F3Net) -\u003e Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images\n\n- [GLNet](https://github.com/Zhu1Teng/GLNet) -\u003e Global to Local: Clip-LSTM-Based Object Detection From Remote Sensing Images\n\n- [SRAF-Net](https://github.com/Complicateddd/SRAF-Net) -\u003e A Scene-Relevant Anchor-Free Object Detection Network in Remote Sensing Images\n\n- [object_detection_in_remote_sensing_images](https://github.com/EEexplorer001/object_detection_in_remote_sensing_images) -\u003e using CNN and attention mechanism\n\n- [SHAPObjectDetection](https://github.com/hiroki-kawauchi/SHAPObjectDetection) -\u003e SHAP-Based Interpretable Object Detection Method for Satellite Imagery\n\n- [NWD](https://github.com/jwwangchn/NWD) -\u003e A Normalized Gaussian Wasserstein Distance for Tiny Object Detection. Uses AI-TOD dataset\n\n- [MSFC-Net](https://github.com/ZhAnGToNG1/MSFC-Net) -\u003e Multiscale Semantic Fusion-Guided Fractal Convolutional Object Detection Network for Optical Remote Sensing Imagery\n\n- [LO-Det](https://github.com/Shank2358/LO-Det) -\u003e LO-Det: Lightweight Oriented Object Detection in Remote Sensing Images\n\n- [R2IPoints](https://github.com/shnew/R2IPoints) -\u003e  Pursuing Rotation-Insensitive Point Representation for Aerial Object Detection\n\n- [Object-Detection](https://github.com/xiaojs18/Object-Detection) -\u003e Multi-Scale Object Detection with the Pixel Attention Mechanism in a Complex Background\n\n- [mmdet-rfla](https://github.com/Chasel-Tsui/mmdet-rfla) -\u003e RFLA: Gaussian Receptive based Label Assignment for Tiny Object Detection\n\n- [Interactive-Multi-Class-Tiny-Object-Detection](https://github.com/ChungYi347/Interactive-Multi-Class-Tiny-Object-Detection) -\u003e Interactive Multi-Class Tiny-Object Detection\n\n- [small-object-detection-benchmark](https://github.com/fcakyon/small-object-detection-benchmark) -\u003e Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection (SAHI)\n\n- [OD-Satellite-iSAID](https://github.com/muzairkhattak/OD-Satellite-iSAID) -\u003e Object Detection in Aerial Images: A Case Study on Performance Improvement using iSAID\n\n- [Large-Selective-Kernel-Network](https://github.com/zcablii/Large-Selective-Kernel-Network) -\u003e Large Selective Kernel Network for Remote Sensing Object Detection\n\n- [Satellite_Imagery_Detection_YOLOV7](https://github.com/Radhika-Keni/Satellite_Imagery_Detection_YOLOV7) -\u003e YOLOV7 applied to xView1 Dataset\n\n- [FSANet](https://github.com/Lausen-Ng/FSANet) -\u003e FSANet: Feature-and-Spatial-Aligned Network for Tiny Object Detection in Remote Sensing Images\n\n- [OAN](https://github.com/Ranchosky/OAN) Fewer is More: Efficient Object Detection in Large Aerial Images, based on MMdetection\n\n- [DOTA-C](https://github.com/hehaodong530/DOTA-C) -\u003e evaluating the robustness of object detection models to 19 types of image quality degradation\n\n- [Satellite-Remote-Sensing-Image-Object-Detection](https://github.com/ypw-lbj/Satellite-Remote-Sensing-Image-Object-Detection) -\u003e using RefineDet \u0026 DOTA dataset\n\n- [yolov5](https://github.com/leticiastachelski/yolov5) -\u003e yolov5 detecting hurricane with Roboflow\n\n- [SFRNet](https://github.com/Ranchosky/SFRNet) -\u003e SFRNet: Fine-Grained Oriented Object Recognition via Separate Feature Refinement\n\n- [contrail-seg](https://github.com/junzis/contrail-seg) -\u003e Neural network models for contrail detection and segmentation\n\n## Object counting\n\nWhen the object count, but not its shape is required, U-net can be used to treat this as an image-to-image translation problem.\n\n- [centroid-unet](https://github.com/gicait/centroid-unet) -\u003e Centroid-UNet is deep neural network model to detect centroids from satellite images\n\n- [cownter_strike](https://github.com/IssamLaradji/cownter_strike) -\u003e counting cows, located with point-annotations, two models: CSRNet (a density-based method) \u0026 LCFCN (a detection-based method)\n\n- [DO-U-Net](https://github.com/ToyahJade/DO-U-Net) -\u003e an effective approach for when the size of an object needs to be known, as well as the number of objects in the image, initially created to segment and count Internally Displaced People (IDP) camps in Afghanistan\n\n- [Cassava Crop Counting](https://medium.com/@wongsirikuln/cassava-standing-crop-counting-869cca486ce3)\n\n- [Counting from Sky](https://github.com/gaoguangshuai/Counting-from-Sky-A-Large-scale-Dataset-for-Remote-Sensing-Object-Counting-and-A-Benchmark-Method) -\u003e A Large-scale Dataset for Remote Sensing Object Counting and A Benchmark Method\n\n- [PSGCNet](https://github.com/gaoguangshuai/PSGCNet) -\u003e PSGCNet: A Pyramidal Scale and Global Context Guided Network for Dense Object Counting in Remote Sensing Images\n\n- [psgcnet](https://github.com/gaoguangshuai/psgcnet) -\u003e A Pyramidal Scale and Global Context Guided Network for Dense Object Counting in Remote-Sensing Images\n\n#\n## Regression\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/regression.png\" width=\"300\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eRegression prediction of windspeed.\u003c/b\u003e\n\u003c/p\u003e\n\nRegression in remote sensing involves predicting continuous variables such as wind speed, tree height, or soil moisture from an image. Both classical machine learning and deep learning approaches can be used to accomplish this task. Classical machine learning utilizes feature engineering to extract numerical values from the input data, which are then used as input for a regression algorithm like linear regression. On the other hand, deep learning typically employs a convolutional neural network (CNN) to process the image data, followed by a fully connected neural network (FCNN) for regression. The FCNN is trained to map the input image to the desired output, providing predictions for the continuous variables of interest. [Image source](https://github.com/h-fuzzy-logic/python-windspeed)\n\n- [python-windspeed](https://github.com/h-fuzzy-logic/python-windspeed) -\u003e Predicting windspeed of hurricanes from satellite images, uses CNN regression in keras\n\n- [hurricane-wind-speed-cnn](https://github.com/23ccozad/hurricane-wind-speed-cnn) -\u003e Predicting windspeed of hurricanes from satellite images, uses CNN regression in keras\n\n- [GEDI-BDL](https://github.com/langnico/GEDI-BDL) -\u003e Global canopy height regression and uncertainty estimation from GEDI LIDAR waveforms with deep ensembles\n\n- [Global-Canopy-Height-Map](https://github.com/AI4Forest/Global-Canopy-Height-Map) -\u003e Estimating Canopy Height at Scale (ICML2024)\n\n- [HighResCanopyHeight](https://github.com/facebookresearch/HighResCanopyHeight) -\u003e code for Meta paper: Very high resolution canopy height maps from RGB imagery using self-supervised vision transformer and convolutional decoder trained on Aerial Lidar\n\n- [Traffic density estimation as a regression problem instead of object detection](https://omdena.com/blog/ai-road-safety/) -\u003e inspired by paper: Traffic density estimation method from small satellite imagery: Towards frequent remote sensing of car traffic\n\n- [OpticalWaveGauging_DNN](https://github.com/OpticalWaveGauging/OpticalWaveGauging_DNN) -\u003e Optical wave gauging using deep neural networks\n\n- [satellite-pose-estimation](https://github.com/eio/satellite-pose-estimation) -\u003e adapts a ResNet50 model architecture to perform pose estimation on several series of satellite images (both real and synthetic)\n\n- [Tropical Cyclone Wind Estimation Competition](https://mlhub.earth/10.34911/rdnt.xs53up) -\u003e on RadiantEarth MLHub\n\n- [DengueNet](https://github.com/mimikuo365/DengueNet-IJCAI) -\u003e DengueNet: Dengue Prediction using Spatiotemporal Satellite Imagery for Resource-Limited Countries\n\n- [tropical_cyclone_uq](https://github.com/nilsleh/tropical_cyclone_uq) -\u003e Uncertainty Aware Tropical Cyclone Wind Speed Estimation from Satellite Data\n\n- [AQNet](https://github.com/CoDIS-Lab/AQNet) -\u003e AQNet - Predicting air quality via multimodal AI and satellite imagery [paper](https://www.sciencedirect.com/science/article/pii/S0034425723001608)\n\n#\n## Cloud detection \u0026 removal\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/clouds.png\" width=\"550\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003e(left) False colour image and (right) a cloud \u0026 shadow mask.\u003c/b\u003e\n\u003c/p\u003e\n\nClouds are a major issue in remote sensing images as they can obscure the underlying ground features. This hinders the accuracy and effectiveness of remote sensing analysis, as the obscured regions cannot be properly interpreted. In order to address this challenge, various techniques have been developed to detect clouds in remote sensing images. Both classical algorithms and deep learning approaches can be employed for cloud detection. Classical algorithms typically use threshold-based techniques and hand-crafted features to identify cloud pixels. However, these techniques can be limited in their accuracy and are sensitive to changes in image appearance and cloud structure. On the other hand, deep learning approaches leverage the power of convolutional neural networks (CNNs) to accurately detect clouds in remote sensing images. These models are trained on large datasets of remote sensing images, allowing them to learn and generalize the unique features and patterns of clouds. The generated cloud mask can be used to identify the cloud pixels and eliminate them from further analysis or, alternatively, cloud inpainting techniques can be used to fill in the gaps left by the clouds. This approach helps to improve the accuracy of remote sensing analysis and provides a clearer view of the ground, even in the presence of clouds. Image adapted from the paper 'Refined UNet Lite: End-to-End Lightweight Network for Edge-precise Cloud Detection'\n\n- [CloudSEN12](https://github.com/cloudsen12) -\u003e Sentinel 2 cloud dataset with a [varierty of models here](https://github.com/cloudsen12/models)\n\n- From [this article on sentinelhub](https://medium.com/sentinel-hub/improving-cloud-detection-with-machine-learning-c09dc5d7cf13) there are three popular classical algorithms that detects thresholds in multiple bands in order to identify clouds. In the same article they propose using semantic segmentation combined with a CNN for a cloud classifier (excellent review paper [here](https://arxiv.org/pdf/1704.06857.pdf)), but state that this requires too much compute resources.\n\n- [This article](https://www.mdpi.com/2072-4292/8/8/666) compares a number of ML algorithms, random forests, stochastic gradient descent, support vector machines, Bayesian method.\n\n- [Segmentation of Clouds in Satellite Images Using Deep Learning](https://medium.com/swlh/segmentation-of-clouds-in-satellite-images-using-deep-learning-a9f56e0aa83d) -\u003e semantic segmentation using a Unet on the Kaggle 38-Cloud dataset\n\n- [Cloud Detection in Satellite Imagery](https://www.azavea.com/blog/2021/02/08/cloud-detection-in-satellite-imagery/) compares FPN+ResNet18 and CheapLab architectures on Sentinel-2 L1C and L2A imagery\n\n- [Benchmarking Deep Learning models for Cloud Detection in Landsat-8 and Sentinel-2 images](https://github.com/IPL-UV/DL-L8S2-UV)\n\n- [Landsat-8 to Proba-V Transfer Learning and Domain Adaptation for Cloud detection](https://github.com/IPL-UV/pvl8dagans)\n\n- [Multitemporal Cloud Masking in Google Earth Engine](https://github.com/IPL-UV/ee_ipl_uv)\n\n- [s2cloudmask](https://github.com/daleroberts/s2cloudmask) -\u003e Sentinel-2 Cloud and Shadow Detection using Machine Learning\n\n- [sentinel2-cloud-detector](https://github.com/sentinel-hub/sentinel2-cloud-detector) -\u003e Sentinel Hub Cloud Detector for Sentinel-2 images in Python\n\n- [dsen2-cr](https://github.com/ameraner/dsen2-cr) -\u003e cloud removal in Sentinel-2 imagery using a deep residual neural network and SAR-optical data fusion, contains the model code, written in Python/Keras, as well as links to pre-trained checkpoints and the SEN12MS-CR dataset\n\n- [pyatsa](https://github.com/agroimpacts/pyatsa) -\u003e Python package implementing the Automated Time-Series Analysis method for masking clouds in satellite imagery developed by Zhu and Helmer 2018\n\n- [decloud](https://github.com/CNES/decloud) -\u003e Decloud enables the training of various deep nets to remove clouds in optical image, using e.g. Sentinel 1 \u0026 2\n\n- [cloudless](https://github.com/BradNeuberg/cloudless) -\u003e Deep learning pipeline for orbital satellite data for detecting clouds\n\n- [Deep-Gapfill](https://github.com/remicres/Deep-Gapfill) -\u003e Official implementation of Optical image gap filling using deep convolutional autoencoder from optical and radar images\n\n- [satellite-cloud-removal-dip](https://github.com/cidcom/satellite-cloud-removal-dip) -\u003e Satellite cloud removal with Deep Image Prior, with [paper](https://www.mdpi.com/2072-4292/14/6/1342)\n\n- [cloudFCN](https://github.com/aliFrancis/cloudFCN) -\u003e Python 3 package for Fully Convolutional Network development, specifically for cloud masking\n\n- [Fmask](https://github.com/GERSL/Fmask) -\u003e Fmask (Function of mask) is used for automated clouds, cloud shadows, snow, and water masking for Landsats 4-9 and Sentinel 2 images, in Matlab. Also see [PyFmask](https://github.com/akalenda/PyFmask)\n\n- [HOW TO USE DEEP LEARNING, PYTORCH LIGHTNING, AND THE PLANETARY COMPUTER TO PREDICT CLOUD COVER IN SATELLITE IMAGERY](https://www.drivendata.co/blog/cloud-cover-benchmark/)\n\n- [cloud-cover-winners](https://github.com/drivendataorg/cloud-cover-winners) -\u003e winning submissions for the On Cloud N: Cloud Cover Detection Challenge\n\n- [On-Cloud-N: Cloud Cover Detection Challenge - 19th Place Solution](https://github.com/max-schaefer-dev/on-cloud-n-19th-place-solution)\n\n- [ukis-csmask](https://github.com/dlr-eoc/ukis-csmask) -\u003e package to masks clouds in Sentinel-2, Landsat-8, Landsat-7 and Landsat-5 images\n\n- [OpenSICDR](https://github.com/dr-lizhiwei/OpenSICDR) -\u003e long list of satellite image cloud detection resources\n\n- [RS-Net](https://github.com/JacobJeppesen/RS-Net) -\u003e  A cloud detection algorithm for satellite imagery based on deep learning\n\n- [Clouds-Segmentation-Project](https://github.com/TamirShalev/Clouds-Segmentation-Project) -\u003e treats as a 3 class problem; Open clouds, Closed clouds and no clouds, uses pytorch on a dataset that consists of IR \u0026 Visual Grayscale images\n\n- [STGAN](https://github.com/ermongroup/STGAN) -\u003e STGAN for Cloud Removal in Satellite Images\n\n- [mcgan-cvprw2017-pytorch](https://github.com/enomotokenji/mcgan-cvprw2017-pytorch) -\u003e Filmy Cloud Removal on Satellite Imagery with Multispectral Conditional Generative Adversarial Nets\n\n- [Cloud-Net: A semantic segmentation CNN for cloud detection](https://github.com/SorourMo/Cloud-Net-A-semantic-segmentation-CNN-for-cloud-detection) -\u003e an end-to-end cloud detection algorithm for Landsat 8 imagery, trained on 38-Cloud Training Set\n\n- [fcd](https://github.com/jnyborg/fcd) -\u003e Fixed-Point GAN for Cloud Detection. A weakly-supervised approach, training with only image-level labels\n\n- [CloudX-Net](https://github.com/sumitkanu/CloudX-Net) -\u003e an efficient and robust architecture used for detection of clouds from satellite images\n\n- [A simple cloud-detection walk-through using Convolutional Neural Network (CNN and U-Net) and fast.ai library](https://medium.com/analytics-vidhya/a-simple-cloud-detection-walk-through-using-convolutional-neural-network-cnn-and-u-net-and-bc745dda4b04)\n\n- [38Cloud-Medium](https://github.com/cordmaur/38Cloud-Medium) -\u003e Walk-through using u-net to detect clouds in satellite images with fast.ai\n\n- [cloud_detection_using_satellite_data](https://github.com/ZhouPeng-NIMST/cloud_detection_using_satellite_data) -\u003e performed on Sentinel 2 data\n\n- [Luojia1-Cloud-Detection](https://github.com/dedztbh/Luojia1-Cloud-Detection) -\u003e Luojia-1 Satellite Visible Band Nighttime Imagery Cloud Detection\n\n- [SEN12MS-CR-TS](https://github.com/PatrickTUM/SEN12MS-CR-TS) -\u003e A Remote Sensing Data Set for Multi-modal Multi-temporal Cloud Removal\n\n- [ES-CCGAN](https://github.com/AnnaCUG/ES-CCGAN) -\u003e This is a dehazed method for remote sensing image, which based on CycleGAN\n\n- [Cloud_Classification_DL](https://github.com/nishp763/Cloud_Classification_DL) -\u003e Classifying cloud organization patterns from satellite images using Deep Learning techniques (Mask R-CNN)\n\n- [CNN-based-Cloud-Detection-Methods](https://github.com/LK-Peng/CNN-based-Cloud-Detection-Methods) -\u003e Understanding the Role of Receptive Field of Convolutional Neural Network for Cloud Detection in Landsat 8 OLI Imagery\n\n- [cloud-removal-deploy](https://github.com/XavierJiezou/cloud-removal-deploy) -\u003e flask app for cloud removal\n\n- [CloudMattingGAN](https://github.com/flyakon/CloudMattingGAN) -\u003e Generative Adversarial Training for Weakly Supervised Cloud Matting\n\n- [km_predict](https://github.com/kappazeta/km_predict) -\u003e KappaMask, or km-predict, is a cloud detector for Sentinel-2 Level-1C and Level-2A input products applied to S2 full image prediction\n\n- [CDnet](https://github.com/nkszjx/CDnet-pytorch-master) -\u003e CNN-Based Cloud Detection for Remote Sensing Imager\n\n- [GLNET](https://github.com/wuchangsheng951/GLNET) -\u003e Convolutional Neural Networks Based Remote Sensing Scene Classification under Clear and Cloudy Environments\n\n- [CDnetV2](https://github.com/nkszjx/CDnetV2-pytorch-master) -\u003e CNN-Based Cloud Detection for Remote Sensing Imagery With Cloud-Snow Coexistence\n\n- [grouped-features-alignment](https://github.com/nkszjx/grouped-features-alignment) -\u003e Unsupervised Domain Adaptation for Cloud Detection Based on Grouped Features Alignment and Entropy Minimization\n\n- [Detecting Cloud Cover Via Sentinel-2 Satellite Data](https://benjaminwarner.dev/2022/03/11/detecting-cloud-cover-via-satellite) -\u003e blog post on Benjamin Warners Top-10 Percent Solution to DrivenData’s On CloudN Competition using fast.ai \u0026 customized version of XResNeXt50. [Repo](https://github.com/warner-benjamin/code_for_blog_posts/tree/main/2022/drivendata_cloudn)\n\n- [AISD](https://github.com/RSrscoder/AISD) -\u003e Deeply supervised convolutional neural network for shadow detection based on a novel aerial shadow imagery dataset\n\n- [CloudGAN](https://github.com/JerrySchonenberg/CloudGAN) -\u003e Detecting and Removing Clouds from RGB-images using Image Inpainting\n\n- [Using GANs to Augment Data for Cloud Image Segmentation Task](https://github.com/jain15mayank/GAN-augmentation-cloud-image-segmentation)\n\n- [Cloud-Segmentation-from-Satellite-Imagery](https://github.com/vedantk-b/Cloud-Segmentation-from-Satellite-Imagery) -\u003e applied to Sentinel-2 dataset\n\n- [HRC_WHU](https://github.com/dr-lizhiwei/HRC_WHU) -\u003e High-Resolution Cloud Detection Dataset comprising 150 RGB images and a resolution varying from 0.5 to 15 m in different global regions\n\n- [MEcGANs](https://github.com/andrzejmizera/MEcGANs) -\u003e Cloud Removal from Satellite Imagery using Multispectral Edge-filtered Conditional Generative Adversarial Networks\n\n- [CloudXNet](https://github.com/shyamfec/CloudXNet) -\u003e CloudX-net: A robust encoder-decoder architecture for cloud detection from satellite remote sensing images\n\n- [cloud-buster](https://github.com/azavea/cloud-buster) -\u003e Sentinel-2 L1C and L2A Imagery with Fewer Clouds\n\n- [SatelliteCloudGenerator](https://github.com/cidcom/SatelliteCloudGenerator) -\u003e A PyTorch-based tool to generate clouds for satellite images\n\n- [SEnSeI](https://github.com/aliFrancis/SEnSeI) -\u003e A python 3 package for developing sensor independent deep learning models for cloud masking in satellite imagery\n\n- [cloud-detection-venus](https://github.com/pesekon2/cloud-detection-venus) -\u003e Using Convolutional Neural Networks for Cloud Detection on VENμS Images over Multiple Land-Cover Types\n\n- [explaining_cloud_effects](https://github.com/JakobCode/explaining_cloud_effects) -\u003e Explaining the Effects of Clouds on Remote Sensing Scene Classification\n\n- [Clouds-Images-Segmentation](https://github.com/DavidHuji/Clouds-Images-Segmentation) -\u003e Marine Stratocumulus Cloud-Type Classification from SEVIRI Using Convolutional Neural Networks\n\n- [DeCloud-GAN](https://github.com/pixiedust18/DeCloud-GAN) -\u003e DeCloud GAN: An Advanced Generative Adversarial Network for Removing Cloud Cover in Optical Remote Sensing Imagery\n\n- [cloud_segmentation_comparative](https://github.com/toelt-llc/cloud_segmentation_comparative) -\u003e BenchCloudVision: A Benchmark Analysis of Deep Learning Approaches for Cloud Detection and Segmentation in Remote Sensing Imagery\n\n- [PLFM-Clouds-Removal](https://github.com/alessandrosebastianelli/PLFM-Clouds-Removal) -\u003e Spatio-Temporal SAR-Optical Data Fusion for Cloud Removal via a Deep Hierarchical Model\n\n- [Cloud-removal-model-collection](https://github.com/littlebeen/Cloud-removal-model-collection) -\u003e A collection of the existing end-to-end cloud removal models\n\n- [SEnSeIv2](https://github.com/aliFrancis/SEnSeIv2) -\u003e Sensor Independent Cloud and Shadow Masking with Ambiguous Labels and Multimodal Inputs\n\n- [cloud-detection-venus](https://github.com/pesekon2/cloud-detection-venus) -\u003e Using Convolutional Neural Networks for Cloud Detection on VENμS Images over Multiple Land-Cover Types\n\n- [UnCRtainTS](https://github.com/PatrickTUM/UnCRtainTS) -\u003e Uncertainty Quantification for Cloud Removal in Optical Satellite Time Series\n\n- [U-TILISE](https://github.com/prs-eth/U-TILISE) -\u003e A Sequence-to-sequence Model for Cloud Removal in Optical Satellite Time Series\n\n#\n## Change detection\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/change.png\" width=\"950\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003e(left) Initial and (middle) after some development, with (right) the change highlighted.\u003c/b\u003e\n\u003c/p\u003e\n\nChange detection is a vital component of remote sensing analysis, enabling the monitoring of landscape changes over time. This technique can be applied to identify a wide range of changes, including land use changes, urban development, coastal erosion, and deforestation. Change detection can be performed on a pair of images taken at different times, or by analyzing multiple images collected over a period of time. It is important to note that while change detection is primarily used to detect changes in the landscape, it can also be influenced by the presence of clouds and shadows. These dynamic elements can alter the appearance of the image, leading to false positives in change detection results. Therefore, it is essential to consider the impact of clouds and shadows on change detection analysis, and to employ appropriate methods to mitigate their influence. [Image source](https://www.mdpi.com/2072-4292/11/3/240)\n\n- [awesome-remote-sensing-change-detection](https://github.com/wenhwu/awesome-remote-sensing-change-detection) lists many datasets and publications\n\n- [Change-Detection-Review](https://github.com/MinZHANG-WHU/Change-Detection-Review) -\u003e A review of change detection methods, including code and open data sets for deep learning\n\n- [STANet](https://github.com/justchenhao/STANet) -\u003eSTANet for remote sensing image change detection\n\n- [UNet-based-Unsupervised-Change-Detection](https://github.com/annabosman/UNet-based-Unsupervised-Change-Detection) -\u003e A convolutional neural network (CNN) and semantic segmentation is implemented to detect the changes between the images, as well as classify the changes into the correct semantic class\n\n- [BIT_CD](https://github.com/justchenhao/BIT_CD) -\u003e Official Pytorch Implementation of Remote Sensing Image Change Detection with Transformers\n\n- [Unstructured-change-detection-using-CNN](https://github.com/vbhavank/Unstructured-change-detection-using-CNN)\n\n- [Siamese neural network to detect changes in aerial images](https://github.com/vbhavank/Siamese-neural-network-for-change-detection) -\u003e uses Keras and VGG16 architecture\n\n- [Change Detection in 3D: Generating Digital Elevation Models from Dove Imagery](https://www.planet.com/pulse/publications/change-detection-in-3d-generating-digital-elevation-models-from-dove-imagery/)\n\n- [QGIS plugin for applying change detection algorithms on high resolution satellite imagery](https://github.com/dymaxionlabs/massive-change-detection)\n\n- [LamboiseNet](https://github.com/hbaudhuin/LamboiseNet) -\u003e Master thesis about change detection in satellite imagery using Deep Learning\n\n- [Fully Convolutional Siamese Networks for Change Detection](https://github.com/rcdaudt/fully_convolutional_change_detection)\n\n- [Urban Change Detection for Multispectral Earth Observation Using Convolutional Neural Networks](https://github.com/rcdaudt/patch_based_change_detection) -\u003e used the Onera Satellite Change Detection (OSCD) dataset\n\n- [IAug_CDNet](https://github.com/justchenhao/IAug_CDNet) -\u003e Official Pytorch Implementation of Adversarial Instance Augmentation for Building Change Detection in Remote Sensing Images\n\n- [dpm-rnn-public](https://github.com/olliestephenson/dpm-rnn-public) -\u003e Code implementing a damage mapping method combining satellite data with deep learning\n\n- [SenseEarth2020-ChangeDetection](https://github.com/LiheYoung/SenseEarth2020-ChangeDetection) -\u003e 1st place solution to the Satellite Image Change Detection Challenge hosted by SenseTime; predictions of five HRNet-based segmentation models are ensembled, serving as pseudo labels of unchanged areas\n\n- [KPCAMNet](https://github.com/I-Hope-Peace/KPCAMNet) -\u003e Python implementation of the paper Unsupervised Change Detection in Multi-temporal VHR Images Based on Deep Kernel PCA Convolutional Mapping Network\n\n- [CDLab](https://github.com/Bobholamovic/CDLab) -\u003e benchmarking deep learning-based change detection methods.\n\n- [Siam-NestedUNet](https://github.com/likyoo/Siam-NestedUNet) -\u003e SNUNet-CD: A Densely Connected Siamese Network for Change Detection of VHR Images\n\n- [SUNet-change_detection](https://github.com/ShaoRuizhe/SUNet-change_detection) -\u003e Implementation of paper SUNet: Change Detection for Heterogeneous Remote Sensing Images from Satellite and UAV Using a Dual-Channel Fully Convolution Network\n\n- [Self-supervised Change Detection in Multi-view Remote Sensing Images](https://github.com/cyx669521/self-supervised_change_detetction)\n\n- [MFPNet](https://github.com/wzjialang/MFPNet) -\u003e Remote Sensing Change Detection Based on Multidirectional Adaptive Feature Fusion and Perceptual Similarity\n\n- [GitHub for the DIUx xView Detection Challenge](https://github.com/DIUx-xView) -\u003e The xView2 Challenge focuses on automating the process of assessing building damage after a natural disaster\n\n- [DASNet](https://github.com/lehaifeng/DASNet) -\u003e Dual attentive fully convolutional siamese networks for change detection of high-resolution satellite images\n\n- [Self-Attention for Raw Optical Satellite Time Series Classification](https://github.com/MarcCoru/crop-type-mapping)\n\n- [planet-movement](https://github.com/rhammell/planet-movement) -\u003e Find and process Planet image pairs to highlight object movement\n\n- [temporal-cluster-matching](https://github.com/microsoft/temporal-cluster-matching) -\u003e detecting change in structure footprints from time series of remotely sensed imagery\n\n- [autoRIFT](https://github.com/nasa-jpl/autoRIFT) -\u003e fast and intelligent algorithm for finding the pixel displacement between two images\n\n- [DSAMNet](https://github.com/liumency/DSAMNet) -\u003e A Deeply Supervised Attention Metric-Based Network and an Open Aerial Image Dataset for Remote Sensing Change Detection\n\n- [SRCDNet](https://github.com/liumency/SRCDNet) -\u003e Super-resolution-based Change Detection Network with Stacked Attention Module for Images with Different Resolutions. SRCDNet is designed to learn and predict change maps from bi-temporal images with different resolutions\n\n- [Land-Cover-Analysis](https://github.com/Kalit31/Land-Cover-Analysis) -\u003e Land Cover Change Detection using Satellite Image Segmentation\n\n- [A deeply supervised image fusion network for change detection in high resolution bi-temporal remote sening images](https://github.com/GeoZcx/A-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images)\n\n- [Satellite-Image-Alignment-Differencing-and-Segmentation](https://github.com/rishi5kesh/Satellite-Image-Alignment-Differencing-and-Segmentation)\n\n- [Change Detection in Multi-temporal Satellite Images](https://github.com/IhebeddineRyahi/Change-detection-in-multitemporal-satellite-images) -\u003e uses Principal Component Analysis (PCA) and K-means clustering\n\n- [Unsupervised Change Detection Algorithm using PCA and K-Means Clustering](https://github.com/leduckhai/Change-Detection-PCA-KMeans) -\u003e in Matlab but has paper\n\n- [ChangeFormer](https://github.com/wgcban/ChangeFormer) -\u003e A Transformer-Based Siamese Network for Change Detection. Uses transformer architecture to address the limitations of CNN in handling multi-scale long-range details. Demonstrates that ChangeFormer captures much finer details compared to the other SOTA methods, achieving better performance on benchmark datasets\n\n- [Heterogeneous_CD](https://github.com/llu025/Heterogeneous_CD) -\u003e Heterogeneous Change Detection in Remote Sensing Images\n\n- [ChangeDetectionProject](https://github.com/previtus/ChangeDetectionProject) -\u003e Trying out Active Learning in with deep CNNs for Change detection on remote sensing data\n\n- [DSFANet](https://github.com/rulixiang/DSFANet) -\u003e Unsupervised Deep Slow Feature Analysis for Change Detection in Multi-Temporal Remote Sensing Images\n\n- [siamese-change-detection](https://github.com/mvkolos/siamese-change-detection) -\u003e Targeted synthesis of multi-temporal remote sensing images for change detection using siamese neural networks\n\n- [Bi-SRNet](https://github.com/ggsDing/Bi-SRNet) -\u003e Bi-Temporal Semantic Reasoning for the Semantic Change Detection in HR Remote Sensing Images\n\n- [SiROC](https://github.com/lukaskondmann/SiROC) -\u003e Spatial Context Awareness for Unsupervised Change Detection in Optical Satellite Images. Applied to Sentinel-2 and high-resolution Planetscope imagery on four datasets\n\n- [DSMSCN](https://github.com/I-Hope-Peace/DSMSCN) -\u003e Tensorflow implementation for Change Detection in Multi-temporal VHR Images Based on Deep Siamese Multi-scale Convolutional Neural Networks\n\n- [RaVAEn](https://github.com/spaceml-org/RaVAEn) -\u003e a lightweight, unsupervised approach for change detection in satellite data based on Variational Auto-Encoders (VAEs) with the specific purpose of on-board deployment. It flags changed areas to prioritise for downlink, shortening the response time\n\n- [SemiCD](https://github.com/wgcban/SemiCD) -\u003e Revisiting Consistency Regularization for Semi-supervised Change Detection in Remote Sensing Images. Achieves the performance of supervised CD even with access to as little as 10% of the annotated training data\n\n- [FCCDN_pytorch](https://github.com/chenpan0615/FCCDN_pytorch) -\u003e FCCDN: Feature Constraint Network for VHR Image Change Detection.\n\n- [INLPG_Python](https://github.com/zcsisiyao/INLPG_Python) -\u003e Structure Consistency based Graph for Unsupervised Change Detection with Homogeneous and Heterogeneous Remote Sensing Images\n\n- [NSPG_Python](https://github.com/zcsisiyao/NSPG_Python) -\u003e Nonlocal patch similarity based heterogeneous remote sensing change detection\n\n- [LGPNet-BCD](https://github.com/TongfeiLiu/LGPNet-BCD) -\u003e Building Change Detection for VHR Remote Sensing Images via Local-Global Pyramid Network and Cross-Task Transfer Learning Strategy\n\n- [DS_UNet](https://github.com/SebastianHafner/DS_UNet) -\u003e Sentinel-1 and Sentinel-2 Data Fusion for Urban Change Detection using a Dual Stream U-Net, uses Onera Satellite Change Detection dataset\n\n- [SiameseSSL](https://github.com/SebastianHafner/SiameseSSL) -\u003e Urban change detection with a Dual-Task Siamese network and semi-supervised learning. Uses SpaceNet 7 dataset\n\n- [CD-SOTA-methods](https://github.com/wgcban/CD-SOTA-methods) -\u003e Remote sensing change detection: State-of-the-art methods and available datasets\n\n- [multimodalCD_ISPRS21](https://github.com/PatrickTUM/multimodalCD_ISPRS21) -\u003e Fusing Multi-modal Data for Supervised Change Detection\n\n- [Unsupervised-CD-in-SITS-using-DL-and-Graphs](https://github.com/ekalinicheva/Unsupervised-CD-in-SITS-using-DL-and-Graphs) -\u003e Unsupervised Change Detection Analysis in Satellite Image Time Series using Deep Learning Combined with Graph-Based Approaches\n\n- [LSNet](https://github.com/qaz670756/LSNet) -\u003e  Extremely Light-Weight Siamese Network For Change Detection in Remote Sensing Image\n\n- [Change-Detection-in-Remote-Sensing-Images](https://github.com/themrityunjay/Change-Detection-in-Remote-Sensing-Images) -\u003e  using PCA \u0026 K-means\n\n- [End-to-end-CD-for-VHR-satellite-image](https://github.com/daifeng2016/End-to-end-CD-for-VHR-satellite-image) -\u003e End-to-End Change Detection for High Resolution Satellite Images Using Improved UNet++\n\n- [Semantic-Change-Detection](https://github.com/daifeng2016/Semantic-Change-Detection) -\u003e SCDNET: A novel convolutional network for semantic change detection in high resolution optical remote sensing imagery\n\n- [ERCNN-DRS_urban_change_monitoring](https://github.com/It4innovations/ERCNN-DRS_urban_change_monitoring) -\u003e Neural Network-Based Urban Change Monitoring with Deep-Temporal Multispectral and SAR Remote Sensing Data\n\n- [EGRCNN](https://github.com/luting-hnu/EGRCNN) -\u003e Edge-guided Recurrent Convolutional Neural Network for Multi-temporal Remote Sensing Image Building Change Detection\n\n- [Unsupervised-Remote-Sensing-Change-Detection](https://github.com/TangXu-Group/Unsupervised-Remote-Sensing-Change-Detection) -\u003e An Unsupervised Remote Sensing Change Detection Method Based on Multiscale Graph Convolutional Network and Metric Learning\n\n- [CropLand-CD](https://github.com/liumency/CropLand-CD) -\u003e A CNN-transformer Network with Multi-scale Context Aggregation for Fine-grained Cropland Change Detection\n\n- [contrastive-surface-image-pretraining](https://github.com/isaaccorley/contrastive-surface-image-pretraining) -\u003e Supervising Remote Sensing Change Detection Models with 3D Surface Semantics\n\n- [dcvaVHROptical](https://github.com/sudipansaha/dcvaVHROptical) -\u003e Unsupervised Deep Change Vector Analysis for Multiple-Change Detection in VHR Images\n\n- [hyperdimensionalCD](https://github.com/sudipansaha/hyperdimensionalCD) -\u003e Change Detection in Hyperdimensional Images Using Untrained Models\n\n- [DSFANet](https://github.com/wwdAlger/DSFANet) -\u003e Unsupervised Deep Slow Feature Analysis for Change Detection in Multi-Temporal Remote Sensing Images\n\n- [FCD-GAN-pytorch](https://github.com/Cwuwhu/FCD-GAN-pytorch) -\u003e Fully Convolutional Change Detection Framework with Generative Adversarial Network (FCD-GAN) is a framework for change detection in multi-temporal remote sensing images\n\n- [DARNet-CD](https://github.com/jimmyli08/DARNet-CD) -\u003e A Densely Attentive Refinement Network for Change Detection Based on Very-High-Resolution Bitemporal Remote Sensing Images\n\n- [xView2_Vulcan](https://github.com/RitwikGupta/xView2-Vulcan) -\u003e Damage assessment using pre and post orthoimagery. Modified + productionized model based off the first-place model from the xView2 challenge.\n\n- [ESCNet](https://github.com/Bobholamovic/ESCNet) -\u003e An End-to-End Superpixel-Enhanced Change Detection Network for Very-High-Resolution Remote Sensing Images\n\n- [ForestCoverChange](https://github.com/annusgit/ForestCoverChange) -\u003e Detecting and Predicting Forest Cover Change in Pakistani Areas Using Remote Sensing Imagery\n\n - [deforestation-detection](https://github.com/vldkhramtsov/deforestation-detection) -\u003e DEEP LEARNING FOR HIGH-FREQUENCY CHANGE DETECTION IN UKRAINIAN FOREST ECOSYSTEM WITH SENTINEL-2\n\n- [forest_change_detection](https://github.com/QuantuMobileSoftware/forest_change_detection) -\u003e forest change segmentation with time-dependent models, including Siamese, UNet-LSTM, UNet-diff, UNet3D models\n\n- [SentinelClearcutDetection](https://github.com/vldkhramtsov/SentinelClearcutDetection) -\u003e Scripts for deforestation detection on the Sentinel-2 Level-A images\n\n- [clearcut_detection](https://github.com/QuantuMobileSoftware/clearcut_detection) -\u003e research \u0026 web-service for clearcut detection\n\n- [CDRL](https://github.com/cjf8899/CDRL) -\u003e Unsupervised Change Detection Based on Image Reconstruction Loss\n\n- [ddpm-cd](https://github.com/wgcban/ddpm-cd) -\u003e  Remote Sensing Change Detection (Segmentation) using Denoising Diffusion Probabilistic Models\n\n- [Remote-sensing-time-series-change-detection](https://github.com/liulianni1688/Remote-sensing-time-series-change-detection) -\u003e Graph-based block-level urban change detection using Sentinel-2 time series\n\n- [austin-ml-change-detection-demo](https://github.com/makepath/austin-ml-change-detection-demo) -\u003e A change detection demo for the Austin area using a pre-trained PyTorch model scaled with Dask on Planet imagery\n\n- [dfc2021-msd-baseline](https://github.com/calebrob6/dfc2021-msd-baseline) -\u003e Multitemporal Semantic Change Detection track of the 2021 IEEE GRSS Data Fusion Competition\n\n- [CorrFusionNet](https://github.com/rulixiang/CorrFusionNet) -\u003e Multi-Temporal Scene Classification and Scene Change Detection with Correlation based Fusion\n\n- [ChangeDetectionPCAKmeans](https://github.com/rulixiang/ChangeDetectionPCAKmeans) -\u003e Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and k-Means Clustering.\n\n- [IRCNN](https://github.com/thebinyang/IRCNN) -\u003e IRCNN: An Irregular-Time-Distanced Recurrent Convolutional Neural Network for Change Detection in Satellite Time Series\n\n- [UTRNet](https://github.com/thebinyang/UTRNet) -\u003e An Unsupervised Time-Distance-Guided Convolutional Recurrent Network for Change Detection in Irregularly Collected Images\n\n- [open-cd](https://github.com/likyoo/open-cd) -\u003e an open source change detection toolbox based on a series of open source general vision task tools\n\n- [Tiny_model_4_CD](https://github.com/AndreaCodegoni/Tiny_model_4_CD) -\u003e TINYCD: A (Not So) Deep Learning Model For Change Detection. Uses LEVIR-CD \u0026 WHU-CD datasets\n\n- [FHD](https://github.com/ZSVOS/FHD) -\u003e Feature Hierarchical Differentiation for Remote Sensing Image Change Detection\n\n- [Change detection with Raster Vision](https://www.azavea.com/blog/2022/04/18/change-detection-with-raster-vision/) -\u003e blog post with Colab notebook\n\n- [building-expansion](https://github.com/reglab/building_expansion) -\u003e Enhancing Environmental Enforcement with Near Real-Time Monitoring: Likelihood-Based Detection of Structural Expansion of Intensive Livestock Farms\n\n- [SaDL_CD](https://github.com/justchenhao/SaDL_CD) -\u003e Semantic-aware Dense Representation Learning for Remote Sensing Image Change Detection\n\n- [EGCTNet_pytorch](https://github.com/chen11221/EGCTNet_pytorch) -\u003e Building Change Detection Based on an Edge-Guided Convolutional Neural Network Combined with a Transformer\n\n- [S2-cGAN](https://git.tu-berlin.de/rsim/S2-cGAN) -\u003e S2-cGAN: Self-Supervised Adversarial Representation Learning for Binary Change Detection in Multispectral Images\n\n- [A-loss-function-for-change-detection](https://github.com/Chuan-shanjia/A-loss-function-for-change-detection) -\u003e UAL: Unchanged Area Loss-Function for Change Detection Networks\n\n- [IEEE_TGRS_SSTFormer](https://github.com/yanhengwang-heu/IEEE_TGRS_SSTFormer) -\u003e Spectral–Spatial–Temporal Transformers for Hyperspectral Image Change Detection\n\n- [DMINet](https://github.com/ZhengJianwei2/DMINet) -\u003e Change Detection on Remote Sensing Images Using Dual-Branch Multilevel Intertemporal Network\n\n- [AFCF3D-Net](https://github.com/wm-Githuber/AFCF3D-Net) -\u003e Adjacent-level Feature Cross-Fusion with 3D CNN for Remote Sensing Image Change Detection\n\n- [DSAHRNet](https://github.com/Githubwujinming/DSAHRNet) -\u003e A Deeply Attentive High-Resolution Network for Change Detection in Remote Sensing Images\n\n- [RDPNet](https://github.com/Chnja/RDPNet) -\u003e RDP-Net: Region Detail Preserving Network for Change Detection\n\n- [BGAAE_CD](https://github.com/xauter/BGAAE_CD) -\u003e Bipartite Graph Attention Autoencoders for Unsupervised Change Detection Using VHR Remote Sensing Images\n\n- [Unsupervised-Change-Detection](https://github.com/voodooed/Unsupervised-Change-Detection) -\u003e Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and k-Means Clustering\n\n- [Metric-CD](https://github.com/wgcban/Metric-CD) -\u003e Deep Metric Learning for Unsupervised Change Detection in Remote Sensing Images\n\n- [HANet-CD](https://github.com/ChengxiHAN/HANet-CD) -\u003e HANet: A hierarchical attention network for change detection with bi-temporal very-high-resolution remote sensing images\n\n- [SRGCAE](https://github.com/ChenHongruixuan/SRGCAE) -\u003e Unsupervised Multimodal Change Detection Based on Structural Relationship Graph Representation Learning\n\n- [change_detection_onera_baselines](https://github.com/previtus/change_detection_onera_baselines) -\u003e Siamese version of U-Net baseline model\n\n- [SiamCRNN](https://github.com/ChenHongruixuan/SiamCRNN) -\u003e Change Detection in Multisource VHR Images via Deep Siamese Convolutional Multiple-Layers Recurrent Neural Network\n\n- [Graph-based methods for change detection in remote sensing images](https://github.com/jfflorez/Graph-based-methods-for-change-detection-in-remote-sensing-images) -\u003e Graph Learning Based on Signal Smoothness Representation for Homogeneous and Heterogeneous Change Detection\n\n- [TransUNetplus2](https://github.com/aj1365/TransUNetplus2) -\u003e TransU-Net++: Rethinking attention gated TransU-Net for deforestation mapping. Uses the Amazon and Atlantic forest dataset\n\n- [AR-CDNet](https://github.com/guanyuezhen/AR-CDNet) -\u003e Towards Accurate and Reliable Change Detection of Remote Sensing Images via Knowledge Review and Online Uncertainty Estimation\n\n- [CICNet](https://github.com/ZhengJianwei2/CICNet) -\u003e Compact Intertemporal Coupling Network for Remote Sensing Change Detection\n\n- [BGINet](https://github.com/JackLiu-97/BGINet) -\u003e Remote Sensing Image Change Detection with Graph Interaction\n\n- [DSNUNet](https://github.com/NightSongs/DSNUNet) -\u003e DSNUNet: An Improved Forest Change Detection Network by Combining Sentinel-1 and Sentinel-2 Images\n\n- [Forest-CD](https://github.com/NightSongs/Forest-CD) -\u003e Forest-CD: Forest Change Detection Network Based on VHR Images\n\n- [S3Net_CD](https://github.com/OMEGA-RS/S3Net_CD) -\u003e Superpixel-Guided Self-Supervised Learning Network for Change Detection in Multitemporal Image Change Detection\n\n- [T-UNet](https://github.com/Pl-2000/T-UNet) -\u003e T-UNet: Triplet UNet for Change Detection in High-Resolution Remote Sensing Images\n\n- [UCDFormer](https://github.com/zhu-xlab/UCDFormer) -\u003e UCDFormer: Unsupervised Change Detection Using a Transformer-driven Image Translation\n\n- [satellite-change-events](https://github.com/utkarshmall13/satellite-change-events) -\u003e Change Event Dataset for Discovery from Spatio-temporal Remote Sensing Imagery, uses Sentinel 2 CaiRoad \u0026 CalFire datasets\n\n- [CACo](https://github.com/utkarshmall13/CACo) -\u003e Change-Aware Sampling and Contrastive Learning for Satellite Images\n\n- [LightCDNet](https://github.com/NightSongs/LightCDNet) -\u003e LightCDNet: Lightweight Change Detection Network Based on VHR Images\n\n- [OpenMineChangeDetection](https://github.com/Dibz15/OpenMineChangeDetection) -\u003e Characterising Open Cast Mining from Satellite Data (Sentinel 2), implements TinyCD, LSNet \u0026 DDPM-CD\n\n- [multi-task-L-UNet](https://github.com/mpapadomanolaki/multi-task-L-UNet) -\u003e A Deep Multi-Task Learning Framework Coupling Semantic Segmentation and Fully Convolutional LSTM Networks for Urban Change Detection. Applied to SpaceNet7 dataset\n\n- [urban_change_detection](https://github.com/SebastianHafner/urban_change_detection) -\u003e Detecting Urban Changes With Recurrent Neural Networks From Multitemporal Sentinel-2 Data. [fabric](https://github.com/granularai/fabric) is another implementation\n\n- [UNetLSTM](https://github.com/mpapadomanolaki/UNetLSTM) -\u003e Detecting Urban Changes With Recurrent Neural Networks From Multitemporal Sentinel-2 Data\n\n- [SDACD](https://github.com/Perfect-You/SDACD) -\u003e An End-to-end Supervised Domain Adaptation Framework for Cross-domain Change Detection\n\n- [CycleGAN-Based-DA-for-CD](https://github.com/pjsoto/CycleGAN-Based-DA-for-CD) -\u003e CycleGAN-based Domain Adaptation for Deforestation Detection\n\n- [CGNet-CD](https://github.com/ChengxiHAN/CGNet-CD) -\u003e Change Guiding Network: Incorporating Change Prior to Guide Change Detection in Remote Sensing Imagery\n\n- [PA-Former](https://github.com/liumency/PA-Former) -\u003e PA-Former: Learning Prior-Aware Transformer for Remote Sensing Building Change Detection\n\n- [AERNet](https://github.com/zjd1836/AERNet) -\u003e AERNet: An Attention-Guided Edge Refinement Network and a Dataset for Remote Sensing Building Change Detection (HRCUS-CD)\n\n- [S1GFlood-Detection](https://github.com/Tamer-Saleh/S1GFlood-Detection) -\u003e DAM-Net: Global Flood Detection from SAR Imagery Using Differential Attention Metric-Based Vision Transformers. Includes S1GFloods dataset\n\n- [Changen](https://github.com/Z-Zheng/Changen) -\u003e Scalable Multi-Temporal Remote Sensing Change Data Generation via Simulating Stochastic Change Process\n\n- [TTP](https://github.com/KyanChen/TTP) -\u003e Time Travelling Pixels: Bitemporal Features Integration with Foundation Model for Remote Sensing Image Change Detection\n\n- [SAM-CD](https://github.com/ggsDing/SAM-CD) -\u003e Adapting Segment Anything Model for Change Detection in HR Remote Sensing Images\n\n- [SCanNet](https://github.com/ggsDing/SCanNet) -\u003e Joint Spatio-Temporal Modeling for Semantic Change Detection in Remote Sensing Images\n\n- [ELGC-Net](https://github.com/techmn/elgcnet) -\u003e Efficient Local-Global Context Aggregation for Remote Sensing Change Detection\n\n- [Official_Remote_Sensing_Mamba](https://github.com/walking-shadow/Official_Remote_Sensing_Mamba) -\u003e RS-Mamba for Large Remote Sensing Image Dense Prediction\n\n- [ChangeMamba](https://github.com/ChenHongruixuan/MambaCD) -\u003e Remote Sensing Change Detection with Spatio-Temporal State Space Model\n\n- [ClearSCD](https://github.com/tangkai-RS/ClearSCD) -\u003e Comprehensively leveraging semantics and change relationships for semantic change detection in high spatial resolution remote sensing imagery\n\n- [RSCaMa](https://github.com/Chen-Yang-Liu/RSCaMa) -\u003e Remote Sensing Image Change Captioning with State Space Model\n\n- [ChangeBind](https://github.com/techmn/changebind) -\u003e A Hybrid Change Encoder for Remote Sensing Change Detection\n\n- [OctaveNet](https://github.com/farhadinima75/OctaveNet) -\u003e An efficient multi-scale pseudo-siamese network for change detection in remote sensing images\n\n- [MaskCD](https://github.com/EricYu97/MaskCD) -\u003e A Remote Sensing Change Detection Network Based on Mask Classification\n\n- [I3PE](https://github.com/ChenHongruixuan/I3PE) -\u003e Exchange means change: an unsupervised single-temporal change detection framework based on intra- and inter-image patch exchange\n\n- [BDANet](https://github.com/ShaneShen/BDANet-Building-Damage-Assessment) -\u003e Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images\n\n- [BAN](https://github.com/likyoo/BAN) -\u003e A New Learning Paradigm for Foundation Model-based Remote Sensing Change Detection\n\n- [ubdd](https://github.com/fzmi/ubdd) -\u003e Learning Efficient Unsupervised Satellite Image-based Building Damage Detection, uses xView2\n\n- [SGSLN](https://github.com/NJU-LHRS/offical-SGSLN) -\u003e Exchanging Dual-Encoder–Decoder: A New Strategy for Change Detection With Semantic Guidance and Spatial Localization\n\n- [ChangeViT](https://github.com/zhuduowang/ChangeViT) -\u003e Unleashing Plain Vision Transformers for Change Detection\n\n- [pytorch-change-models](https://github.com/Z-Zheng/pytorch-change-models) -\u003e out-of-box contemporary spatiotemporal change model implementations, standard metrics, and datasets\n\n- [FFCTL](https://github.com/lauraset/FFCTL) -\u003e A full-level fused cross-task transfer learning method for building change detection using noise-robust pretrained networks on crowdsourced labels\n\n- [SARAS-Net](https://github.com/f64051041/SARAS-Net) -\u003e SARAS-Net: Scale And Relation Aware Siamese Network for Change Detection\n\n- [Change_Detection_FCNs](https://github.com/DLoboT/Change_Detection_FCNs) -\u003e Deforestation Detection with Fully Convolutional Networks in the Amazon Forest from Landsat-8 and Sentinel-2 Images\n\n- [HyperNet](https://github.com/meiqihu/HyperNet) -\u003e HyperNet: Self-Supervised Hyperspectral SpatialSpectral Feature Understanding Network for Hyperspectral Change Detection [paper](https://ieeexplore.ieee.org/document/9934933)\n\n- [CMCDNet](https://github.com/CAU-HE/CMCDNet) -\u003e CMCDNet: Cross-modal change detection flood extraction based on convolutional neural network [paper](https://www.sciencedirect.com/science/article/pii/S1569843223000195)\n\n- [Dsfer-Net](https://github.com/ShizhenChang/Dsfer-Net) -\u003e A Deep Supervision and Feature Retrieval Network for Bitemporal Change Detection Using Modern Hopfield Network [paper](https://arxiv.org/pdf/2304.01101)\n\n- [Simple-Remote-Sensing-Change-Detection-Framework](https://github.com/walking-shadow/Simple-Remote-Sensing-Change-Detection-Framework) -\u003e Simplified implementation of remote sensing change detection based on Pytorch\n\n- [BCE-Net](https://github.com/liaochengcsu/BCE-Net) -\u003e BCE-Net: Reliable Building Footprints Change Extraction based on Historical Map and Up-to-Date Images using Contrastive Learning\n\n- [sits-change-detection](https://github.com/adebowaledaniel/sits-change-detection) -\u003e Detecting Land Cover Changes Between Satellite Image Time Series By Exploiting Self-Supervised Representation Learning Capabilities\n\n- [USSFC-Net](https://github.com/SUST-reynole/USSFC-Net) -\u003e Ultralightweight Spatial–Spectral Feature Cooperation Network for Change Detection in Remote Sensing Images [paper](https://ieeexplore.ieee.org/document/10081023)\n\n- [VcT_Remote_Sensing_Change_Detection](https://github.com/Event-AHU/VcT_Remote_Sensing_Change_Detection) -\u003e VcT: Visual change Transformer for Remote Sensing Image Change Detection [IEEE](https://ieeexplore.ieee.org/document/10294300)\n\n#\n## Time series\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/time-series.png\" width=\"350\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003ePrediction of the next image in a series.\u003c/b\u003e\n\u003c/p\u003e\n\nThe analysis of time series observations in remote sensing data has numerous applications, including enhancing the accuracy of classification models and forecasting future patterns and events. [Image source](https://www.mdpi.com/2072-4292/13/23/4822). Note: since classifying crops and predicting crop yield are such prominent use case for time series data, these tasks have dedicated sections after this one.\n\n- [LANDSAT Time Series Analysis for Multi-temporal Land Cover Classification using Random Forest](https://github.com/agr-ayush/Landsat-Time-Series-Analysis-for-Multi-Temporal-Land-Cover-Classification)\n\n- [temporalCNN](https://github.com/charlotte-pel/temporalCNN) -\u003e Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series\n\n- [pytorch-psetae](https://github.com/VSainteuf/pytorch-psetae) -\u003e Satellite Image Time Series Classification with Pixel-Set Encoders and Temporal Self-Attention\n\n- [satflow](https://github.com/openclimatefix/satflow) -\u003e optical flow models for predicting future satellite images from current and past ones\n\n- [esa-superresolution-forecasting](https://github.com/PiSchool/esa-superresolution-forecasting) -\u003e Forecasting air pollution using ESA Sentinel-5p data, and an encoder-decoder convolutional LSTM neural network architecture\n\n- [lightweight-temporal-attention-pytorch](https://github.com/VSainteuf/lightweight-temporal-attention-pytorch) -\u003e Light Temporal Attention Encoder (L-TAE) for satellite image time series\n\n- [dtwSat](https://github.com/vwmaus/dtwSat) -\u003e Time-Weighted Dynamic Time Warping for satellite image time series analysis\n\n- [MTLCC](https://github.com/MarcCoru/MTLCC) -\u003e Multitemporal Land Cover Classification Network. A recurrent neural network approach to encode multi-temporal data for land cover classification\n\n- [PWWB](https://github.com/PannuMuthu/PWWB) -\u003e Real-Time Spatiotemporal Air Pollution Prediction with Deep Convolutional LSTM through Satellite Image Analysis\n\n- [spaceweather](https://github.com/sarttiso/spaceweather) -\u003e predicting geomagnetic storms from satellite measurements of the solar wind and solar corona, uses LSTMs\n\n- [ConvTimeLSTM](https://github.com/jdiaz4302/ConvTimeLSTM) -\u003e Extension of ConvLSTM and Time-LSTM for irregularly spaced images, appropriate for Remote Sensing\n\n- [dl-time-series](https://github.com/NexGenMap/dl-time-series) -\u003e Deep Learning algorithms applied to characterization of Remote Sensing time-series\n\n- [tpe](https://github.com/jnyborg/tpe) -\u003e Generalized Classification of Satellite Image Time Series With Thermal Positional Encoding\n\n- [wildfire_forecasting](https://github.com/Orion-AI-Lab/wildfire_forecasting) -\u003e Deep Learning Methods for Daily Wildfire Danger Forecasting. Uses ConvLSTM\n\n- [satellite_image_forecasting](https://github.com/rudolfwilliam/satellite_image_forecasting) -\u003e predict future satellite images from past ones using features such as precipitation and elevation maps. Entry for the [EarthNet2021](https://www.earthnet.tech/) challenge\n\n- [Deep Learning for Cloud Gap-Filling on Normalized Difference Vegetation Index using Sentinel Time-Series](https://github.com/Agri-Hub/Deep-Learning-for-Cloud-Gap-Filling-on-Normalized-Difference-Vegetation-Index) -\u003e A CNN-RNN based model that identifies correlations between optical and SAR data and exports dense Normalized Difference Vegetation Index (NDVI) time-series of a static 6-day time resolution and can be used for Events Detection tasks\n\n- [DeepSatModels](https://github.com/michaeltrs/DeepSatModels) -\u003e ViTs for SITS: Vision Transformers for Satellite Image Time Series\n\n- [Presto](https://github.com/nasaharvest/presto) -\u003e Lightweight, Pre-trained Transformers for Remote Sensing Timeseries\n\n- [LULC mapping using time series data \u0026 spectral bands](https://github.com/developmentseed/time-series-for-lulc) -\u003e uses 1D convolutions that learn from time-series data. Accompanies blog post: [Time-Traveling Pixels: A Journey into Land Use Modeling](https://developmentseed.org/blog/2023-06-29-time-travel-pixels)\n\n- [hurricane-net](https://github.com/hammad93/hurricane-net) -\u003e A deep learning framework for forecasting Atlantic hurricane trajectory and intensity.\n\n- [CAPES](https://github.com/twin22jw/CAPES/tree/main) -\u003e Construction changes are detected using the U-net model and satellite time series\n\n- [Exchanger4SITS](https://github.com/TotalVariation/Exchanger4SITS) -\u003e Rethinking the Encoding of Satellite Image Time Series\n\n- [Rapid Wildfire Hotspot Detection Using Self-Supervised Learning on Temporal Remote Sensing Data](https://github.com/links-ads/igarss-multi-temporal-hotspot-detection)\n\n- [stenn-pytorch](https://github.com/ThinkPak/stenn-pytorch) -\u003e A Spatio-temporal Encoding Neural Network for Semantic Segmentation of Satellite Image Time Series\n\n- [SITS-Former](https://github.com/linlei1214/SITS-Former) -\u003e SITS-Former: A Pre-Trained Spatio-Spectral-Temporal Representation Model for Sentinel-2 Time Series Classification\n\n- [graph-dynamic-earth-net](https://github.com/corentin-dfg/graph-dynamic-earth-net) -\u003e Graph Dynamic Earth Net: Spatio-Temporal Graph Benchmark for Satellite Image Time Series [paper](https://ieeexplore.ieee.org/abstract/document/10281458)\n\n- [multi-stage-convSTAR-network](https://github.com/0zgur0/multi-stage-convSTAR-network) -\u003e Pytorch implementation for hierarchical time series classification with multi-stage convolutional RNN [paper](https://arxiv.org/pdf/2102.08820.pdf)\n\n#\n## Crop classification\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/crops.jpg\" width=\"600\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003e(left) false colour image and (right) the crop map.\u003c/b\u003e\n\u003c/p\u003e\n\nCrop classification in remote sensing is the identification and mapping of different crops in images or sequences of images. It aims to provide insight into the distribution and composition of crops in a specific area, with applications that include monitoring crop growth and evaluating crop damage. Both traditional machine learning methods, such as decision trees and support vector machines, and deep learning techniques, such as convolutional neural networks (CNNs), can be used to perform crop classification. The optimal method depends on the size and complexity of the dataset, the desired accuracy, and the available computational resources. However, the success of crop classification relies heavily on the quality and resolution of the input data, as well as the availability of labeled training data. Image source: High resolution satellite imaging sensors for precision agriculture by Chenghai Yang\n\n- [Classification of Crop Fields through Satellite Image Time Series](https://medium.com/dida-machine-learning/classification-of-crop-fields-through-satellite-image-time-serie-dida-machine-learning-9b64ce2b8c10) -\u003e using a [pytorch-psetae](https://github.com/VSainteuf/pytorch-psetae) \u0026 Sentinel-2 data\n\n- [CropDetectionDL](https://github.com/karimmamer/CropDetectionDL) -\u003e using GRU-net, First place solution for Crop Detection from Satellite Imagery competition organized by CV4A workshop at ICLR 2020\n\n- [Radiant-Earth-Spot-the-Crop-Challenge](https://github.com/DariusTheGeek/Radiant-Earth-Spot-the-Crop-Challenge) -\u003e The main objective of this challenge was to use time-series of Sentinel-2 multi-spectral data to classify crops in the Western Cape of South Africa. The challenge was to build a machine learning model to predict crop type classes for the test dataset\n\n- [Crop-Classification](https://github.com/bhavesh907/Crop-Classification) -\u003e crop classification using multi temporal satellite images\n\n- [DeepCropMapping](https://github.com/Lab-IDEAS/DeepCropMapping) -\u003e A multi-temporal deep learning approach with improved spatial generalizability for dynamic corn and soybean mapping, uses LSTM\n\n- [CropMappingInterpretation](https://github.com/Lab-IDEAS/CropMappingInterpretation) -\u003e An interpretation pipeline towards understanding multi-temporal deep learning approaches for crop mapping\n\n- [timematch](https://github.com/jnyborg/timematch) -\u003e A method to perform unsupervised cross-region adaptation of crop classifiers trained with satellite image time series. We also introduce an open-access dataset for cross-region adaptation with SITS from four different regions in Europe\n\n- [elects](https://github.com/MarcCoru/elects) -\u003e End-to-End Learned Early Classification of Time Series for In-Season Crop Type Mapping\n\n- [3d-fpn-and-time-domain](https://gitlab.com/ignazio.gallo/sentinel-2-time-series-with-3d-fpn-and-time-domain-cai) -\u003e Sentinel 2 Time Series Analysis with 3D Feature Pyramid Network and Time Domain Class Activation Intervals for Crop Mapping\n\n- [in-season-and-dynamic-crop-mapping](https://gitlab.com/artelabsuper/in-season-and-dynamic-crop-mapping) -\u003e In-season and dynamic crop mapping using 3D convolution neural networks and sentinel-2 time series, uses the Lombardy crop dataset\n\n- [MultiviewCropClassification](https://github.com/fmenat/MultiviewCropClassification) -\u003e A COMPARATIVE ASSESSMENT OF MULTI-VIEW FUSION LEARNING FOR CROP CLASSIFICATION\n\n- [Detection of manure application on crop fields leveraging satellite data and Machine Learning](https://github.com/Amatofrancesco99/master-thesis)\n\n- [StressNet: A spatial-spectral-temporal deformable attention-based framework for water stress classification in maize](https://github.com/tejasri19/Stressnet) -\u003e Water Stress Classification on Multispectral data of Maize captured by UAV\n\n- [XAI4EO](https://github.com/adelabbs/XAI4EO) -\u003e Towards Explainable AI4EO: an explainable DL approach for crop type mapping using SITS\n\n- [model_ecaas_agrifieldnet_gold](https://github.com/radiantearth/model_ecaas_agrifieldnet_gold) -\u003e AgriFieldNet Model for Crop Types Detection. First place solution of the of the [Zindi AgriFieldNet India Challenge](https://zindi.africa/competitions/agrifieldnet-india-challenge) for Crop Types Detection from Satellite Imagery. \n\n#\n## Crop yield \u0026 vegetation forecasting\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/yield.png\" width=\"600\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eWheat yield data. Blue vertical lines denote observation dates.\u003c/b\u003e\n\u003c/p\u003e\n\nCrop yield is a crucial metric in agriculture, as it determines the productivity and profitability of a farm. It is defined as the amount of crops produced per unit area of land and is influenced by a range of factors including soil fertility, weather conditions, the type of crop grown, and pest and disease control. By utilizing time series of satellite images, it is possible to perform accurate crop type classification and take advantage of the seasonal variations specific to certain crops. This information can be used to optimize crop management practices and ultimately improve crop yield. However, to achieve accurate results, it is essential to consider the quality and resolution of the input data, as well as the availability of labeled training data. Appropriate pre-processing and feature extraction techniques must also be employed. [Image source](https://www.mdpi.com/2072-4292/14/17/4193).\n\n- [Crop yield Prediction with Deep Learning](https://github.com/JiaxuanYou/crop_yield_prediction) -\u003e Deep Gaussian Process for Crop Yield Prediction Based on Remote Sensing Data\n\n- [Deep-Transfer-Learning-Crop-Yield-Prediction](https://github.com/sustainlab-group/Deep-Transfer-Learning-Crop-Yield-Prediction)\n\n- [Crop-Yield-Prediction-using-ML](https://github.com/VaibhavSaini19/Crop-Yield-Prediction-using-ML) -\u003e A simple Web application developed in order to provide the farmers/users an approximation on how much amount of crop yield will be produced depending upon the given input\n\n- [Building a Crop Yield Prediction App in Senegal Using Satellite Imagery and Jupyter Voila](https://omdena.com/blog/yield-prediction/)\n\n- [Crop Yield Prediction Using Deep Neural Networks and LSTM](https://omdena.com/blog/deep-learning-yield-prediction/)\n\n- [Deep transfer learning techniques for crop yield prediction, published in COMPASS 2018](https://github.com/AnnaXWang/deep-transfer-learning-crop-prediction)\n\n- [Understanding crop yield predictions from CNNs](https://github.com/brad-ross/crop-yield-prediction-project)\n\n- [Advanced Deep Learning Techniques for Predicting Maize Crop Yield using Sentinel-2 Satellite Imagery](https://zionayomide.medium.com/advanced-deep-learning-techniques-for-predicting-maize-crop-yield-using-sentinel-2-satellite-1b63ac8b0789)\n\n- [pycrop-yield-prediction](https://github.com/gabrieltseng/pycrop-yield-prediction) -\u003e Deep Gaussian Process for Crop Yield Prediction\n\n- [PredictYield](https://github.com/dberm312/PredictYield) -\u003e using data scraped from Google Earth Engine, this predicts the yield of Corn, Soybean, and Wheat in the USA with Keras\n\n- [Crop-Yield-Prediction-and-Estimation-using-Time-series-remote-sensing-data](https://github.com/mahimatendulkar/Crop-Yield-Prediction-and-Estimation-using-Time-series-remote-sensing-data.)\n\n- [Yield-Prediction-Using-Sentinel-Data](https://github.com/meet-sapu/Crop-Yield-Prediction-Using-Satellite-Imagery)\n\n- [SPACY](https://github.com/rlee360/PLaTYPI) -\u003e Satellite Prediction of Aggregate Corn Yield\n\n- [cropyieldArticle](https://github.com/myliheik/cropyieldArticle) -\u003e Scalable Crop Yield Prediction with Sentinel-2 Time Series and Temporal Convolutional Network\n\n- [CNN-RNN-Yield-Prediction](https://github.com/saeedkhaki92/CNN-RNN-Yield-Prediction) -\u003eA CNN-RNN Framework for Crop Yield Prediction\n\n- [Yield-Prediction-DNN](https://github.com/saeedkhaki92/Yield-Prediction-DNN) -\u003e Crop Yield Prediction Using Deep Neural Networks\n\n- [MMST-ViT](https://github.com/fudong03/MMST-ViT) -\u003e MMST-ViT: Climate Change-aware Crop Yield Prediction via Multi-Modal Spatial-Temporal Vision Transformer. This paper utilizes the Tiny CropNet dataset\n\n- [Greenearthnet](https://github.com/vitusbenson/greenearthnet) -\u003e Multi-modal learning for geospatial vegetation forecasting\n\n- [crop-forecasting](https://github.com/association-rosia/crop-forecasting) -\u003e Predicting rice field yields\n\n- [SICKLE](https://github.com/Depanshu-Sani/SICKLE) -\u003e A Multi-Sensor Satellite Imagery Dataset Annotated with Multiple Key Cropping Parameters. Basline solutions: U-TAE, U-Net3D and ConvLSTM\n\n- [yieldCNN](https://github.com/waldnerf/yieldCNN) -\u003e Training temporal Convolution Neural Networks (CNNs) on satellite image time series for yield forecasting\n\n#\n## Wealth and economic activity\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/economic.png\" width=\"450\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eCOVID-19 impacts on human and economic activities.\u003c/b\u003e\n\u003c/p\u003e\n\nThe traditional approach of collecting economic data through ground surveys is a time-consuming and resource-intensive process. However, advancements in satellite technology and machine learning offer an alternative solution. By utilizing satellite imagery and applying machine learning algorithms, it is possible to obtain accurate and current information on economic activity with greater efficiency. This shift towards satellite imagery-based forecasting not only provides cost savings but also offers a wider and more comprehensive perspective of economic activity. As a result, it is poised to become a valuable asset for both policymakers and businesses. [Image source](https://arxiv.org/abs/2004.07438).\n\n- [Using publicly available satellite imagery and deep learning to understand economic well-being in Africa, Nature Comms 22 May 2020](https://www.nature.com/articles/s41467-020-16185-w) -\u003e Used CNN on Ladsat imagery (night \u0026 day) to predict asset wealth of African villages\n\n- [satellite_led_liverpool](https://github.com/darribas/satellite_led_liverpool) -\u003e  Remote Sensing-Based Measurement of Living Environment Deprivation - Improving Classical Approaches with Machine Learning\n\n- [Predicting_Energy_Consumption_With_Convolutional_Neural_Networks](https://github.com/healdz/Predicting_Energy_Consumption_With_Convolutional_Neural_Networks)\n\n- [SustainBench](https://github.com/sustainlab-group/sustainbench/) -\u003e Benchmarks for Monitoring the Sustainable Development Goals with Machine Learning\n\n- [Measuring the Impacts of Poverty Alleviation Programs with Satellite Imagery and Deep Learning](https://github.com/luna983/beyond-nightlight)\n\n- [deeppop](https://deeppop.github.io/) -\u003e Deep Learning Approach for Population Estimation from Satellite Imagery, also [on Github](https://github.com/deeppop)\n\n- [Estimating telecoms demand in areas of poor data availability](https://github.com/edwardoughton/taddle)\n\n- [satimage](https://github.com/mani-shailesh/satimage) -\u003e Code and models for the manuscript \"Predicting Poverty and Developmental Statistics from Satellite Images using Multi-task Deep Learning\". Predict the main material of a roof, source of lighting and source of drinking water for properties, from satellite imagery\n\n- [africa_poverty](https://github.com/sustainlab-group/africa_poverty) -\u003e Using publicly available satellite imagery and deep learning to understand economic well-being in Africa\n\n- [Predicting-Poverty](https://github.com/jmather625/predicting-poverty-replication) -\u003e Combining satellite imagery and machine learning to predict poverty, in PyTorch\n\n- [income-prediction](https://github.com/tnarayanan/income-prediction) -\u003e Predicting average yearly income based on satellite imagery using CNNs, uses pytorch\n\n- [urban_score](https://github.com/Sungwon-Han/urban_score) -\u003e Learning to score economic development from satellite imagery\n\n- [READ](https://github.com/Sungwon-Han/READ) -\u003e Lightweight and robust representation of economic scales from satellite imagery\n\n - [Slum-classification](https://github.com/Jesse-DE/Slum-classification) -\u003e Binary classification on a very high-resolution satellite image in case of mapping informal settlements using unet\n\n - [Predicting_Poverty](https://github.com/cyuancheng/Predicting_Poverty) -\u003e uses daytime \u0026 luminosity of nighttime satellite images\n\n- [Cancer-Prevalence-Satellite-Images](https://github.com/theJamesChen/Cancer-Prevalence-Satellite-Images) -\u003e Predict Health Outcomes from Features of Satellite Images\n\n- [Mapping Poverty in Bangladesh with Satellite Images and Deep Learning](https://github.com/huydang90/Mapping-Poverty-With-Satellite-Images) -\u003e combines health data with OpenStreetMaps Data \u0026 night and daytime satellite imagery\n\n - [Population Estimation from Satellite Imagery](https://github.com/ManuelSerranoR/Population-Estimation-from-Satellite-Imagery-using-Deep-Learning)\n\n- [Deep_Learning_Satellite_Imd](https://github.com/surendran-berkeley/Deep_Learning_Satellite_Imd) -\u003e Using Deep Learning on Satellite Imagery to predict population and economic indicators\n\n#\n## Disaster response\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/disaster.png\" width=\"750\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eDetecting buildings destroyed in a disaster.\u003c/b\u003e\n\u003c/p\u003e\n\nRemote sensing images are used in disaster response to identify and assess damage to an area. This imagery can be used to detect buildings that are damaged or destroyed, identify roads and road networks that are blocked, determine the size and shape of a disaster area, and identify areas that are at risk of flooding. Remote sensing images can also be used to detect and monitor the spread of forest fires and monitor vegetation health. Also checkout the sections on change detection and water/fire/building segmentation. [Image source](https://developer.nvidia.com/blog/ai-helps-detect-disaster-damage-from-satellite-imagery/).\n\n- [DisaVu](https://github.com/SrzStephen/DisaVu) -\u003e combines building \u0026 damage detection and provides an app for viewing predictions\n\n- [Soteria](https://github.com/Soteria-ai/Soteria) -\u003e uses machine learning with satellite imagery to map natural disaster impacts for faster emergency response\n\n- [DisasterHack](https://github.com/MarjorieRWillner/DisasterHack) -\u003e Wildfire Mitigation: Computer Vision Identification of Hazard Fuels Using Landsat\n\n- [forestcasting](https://github.com/ivanzvonkov/forestcasting) -\u003e Forest fire prediction powered by analytics\n\n- [Machine Learning-based Damage Assessment for Disaster Relief on Google AI blog](https://ai.googleblog.com/2020/06/machine-learning-based-damage.html) -\u003e uses object detection to locate buildings, then a classifier to determine if a building is damaged. Challenge of generalising due to small dataset\n\n- [hurricane_damage](https://github.com/allankapoor/hurricane_damage) -\u003e Post-hurricane structure damage assessment based on aerial imagery with CNN\n\n- [rescue](https://github.com/dbdmg/rescue) -\u003e code of the paper: Attention to fires: multi-channel deep-learning models forwildfire severity prediction\n\n-. [Disaster-Classification](https://github.com/bostankhan6/Disaster-Classification) -\u003e A disaster classification model to predict the type of disaster given an input image\n\n- [Coarse-to-fine weakly supervised learning method for green plastic cover segmentation](https://github.com/lauraset/Coarse-to-fine-weakly-supervised-GPC-segmentation)\n\n- [Detection of destruction in satellite imagery](https://github.com/usmanali414/Destruction-Detection-in-Satellite-Imagery)\n\n- [BDD-Net](https://github.com/jinyuan30/Recognize-damaged-buildings) -\u003e A General Protocol for Mapping Buildings Damaged by a Wide Range of Disasters Based on Satellite Imagery\n\n- [building-segmentation-disaster-resilience](https://github.com/kbrodt/building-segmentation-disaster-resilience) -\u003e 2nd place solution in the Open Cities AI Challenge: Segmenting Buildings for Disaster Resilience\n\n- [Flooding Damage Detection from Post-Hurricane Satellite Imagery Based on Convolutional Neural Networks](https://github.com/weining20000/Flooding-Damage-Detection-from-Post-Hurricane-Satellite-Imagery-Based-on-CNN)\n\n- [IBM-Disaster-Response-Hack](https://github.com/NicoDeshler/IBM-Disaster-Response-Hack) -\u003e identifying optimal terrestrial routes through calamity-stricken areas. Satellite image data informs road condition assessment and obstruction detection\n\n- [Automatic Damage Annotation on Post-Hurricane Satellite Imagery](https://dds-lab.github.io/disaster-damage-detection/) -\u003e detect damaged buildings using tensorflow object detection API. With repos [here](https://github.com/DDS-Lab/disaster-image-processing) and [here](https://github.com/annieyan/PreprocessSatelliteImagery-ObjectDetection)\n\n- [Hurricane-Damage-Detection](https://github.com/Ryan-Awad/Hurricane-Damage-Detection) -\u003e Waterloo's Hack the North 2020++ submission. A convolutional neural network model used to detect hurricane damage in RGB satellite images\n\n- [wildfire_forecasting](https://github.com/Orion-AI-Lab/wildfire_forecasting) -\u003e Deep Learning Methods for Daily Wildfire Danger Forecasting. Uses ConvLSTM\n\n- [Satellite Image Analysis with fast.ai for Disaster Recovery](https://appsilon.com/satellite-image-analysis-with-fast-ai-for-disaster-recovery/)\n\n- [shackleton](https://github.com/avanetten/shackleton) -\u003e leverages remote sensing imagery and machine learning techniques to provide insights into various transportation and evacuation scenarios in an interactive dashboard that conducts real-time computation\n\n- [ai-vegetation-fuel](https://github.com/ecmwf-projects/ai-vegetation-fuel) -\u003e Predicting Fuel Load from earth observation data using Machine Learning, using LightGBM \u0026 CatBoost\n\n- [AI Helps Detect Disaster Damage From Satellite Imagery](https://developer.nvidia.com/blog/ai-helps-detect-disaster-damage-from-satellite-imagery/) -\u003e NVIDIA blog post\n\n- [Turkey-Earthquake-2023-Building-Change-Detection](https://github.com/blackshark-ai/Turkey-Earthquake-2023-Building-Change-Detection) -\u003e The repository contains building footprints derived from Maxar open data imagery and change detection results by blackshark-ai\n\n- [MS4D-Net-Building-Damage-Assessment](https://github.com/YJ-He/MS4D-Net-Building-Damage-Assessment) -\u003e MS4D-Net: Multitask-Based Semi-Supervised Semantic Segmentation Framework with Perturbed Dual Mean Teachers for Building Damage Assessment from High-Resolution Remote Sensing Imagery\n\n- [DAHiTra](https://github.com/nka77/DAHiTra) -\u003e Large-scale Building Damage Assessment using a Novel Hierarchical Transformer Architecture on Satellite Images. Uses xView2 xBD dataset\n\n- [skai](https://github.com/google-research/skai) -\u003e a machine learning based tool from Goolge for performing automatic building damage assessments on aerial imagery of disaster sites.\n\n- [building-damage-assessment](https://github.com/microsoft/building-damage-assessment) -\u003e A toolkit that enables building damage assessments from remotely sensed imagery\n\n- [building-damage-assessment-cnn-siamese](https://github.com/microsoft/building-damage-assessment-cnn-siamese) -\u003e from the Microsoft Ai for Good lab\n\n#\n## Super-resolution\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/super-res.jpg\" width=\"500\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eSuper resolution using multiple low resolution images as input.\u003c/b\u003e\n\u003c/p\u003e\n\nSuper-resolution is a technique aimed at improving the resolution of an imaging system. This process can be applied prior to other image processing steps to increase the visibility of small objects or boundaries. Despite its potential benefits, the use of super-resolution is controversial due to the possibility of introducing artifacts that could be mistaken for real features. Super-resolution techniques are broadly categorized into two groups: single image super-resolution (SISR) and multi-image super-resolution (MISR). SISR focuses on enhancing the resolution of a single image, while MISR utilizes multiple images of the same scene to create a high-resolution output. Each approach has its own advantages and limitations, and the choice of method depends on the specific application and desired outcome. [Image source](https://github.com/worldstrat/worldstrat).\n\n### Multi image super-resolution (MISR)\nNote that nearly all the MISR publications resulted from the [PROBA-V Super Resolution competition](https://kelvins.esa.int/proba-v-super-resolution/)\n\n- [deepsum](https://github.com/diegovalsesia/deepsum) -\u003e Deep neural network for Super-resolution of Unregistered Multitemporal images (ESA PROBA-V challenge)\n\n- [3DWDSRNet](https://github.com/frandorr/3DWDSRNet) -\u003e Satellite Image Multi-Frame Super Resolution (MISR) Using 3D Wide-Activation Neural Networks\n\n- [RAMS](https://github.com/EscVM/RAMS) -\u003e Multi-Image Super Resolution of Remotely Sensed Images Using Residual Attention Deep Neural Networks\n\n- [TR-MISR](https://github.com/Suanmd/TR-MISR) -\u003e  Transformer-based MISR framework for the the PROBA-V super-resolution challenge. With [paper](https://ieeexplore.ieee.org/abstract/document/9684717)\n\n- [HighRes-net](https://github.com/ElementAI/HighRes-net) -\u003e Pytorch implementation of HighRes-net, a neural network for multi-frame super-resolution, trained and tested on the European Space Agency’s Kelvin competition\n\n- [ProbaVref](https://github.com/centreborelli/ProbaVref) -\u003e Repurposing the Proba-V challenge for reference-aware super resolution\n\n- [MSTT-STVSR](https://github.com/XY-boy/MSTT-STVSR) -\u003e Space-time Super-resolution for Satellite Video: A Joint Framework Based on Multi-Scale Spatial-Temporal Transformer, JAG, 2022\n\n- [Self-Supervised Super-Resolution for Multi-Exposure Push-Frame Satellites](https://centreborelli.github.io/HDR-DSP-SR/)\n\n- [DDRN](https://github.com/kuijiang94/DDRN) -\u003e Deep Distillation Recursive Network for Video Satellite Imagery Super-Resolution\n\n-[worldstrat](https://github.com/worldstrat/worldstrat) -\u003e SISR and MISR implementations of SRCNN\n\n- [MISR-GRU](https://github.com/rarefin/MISR-GRU) -\u003e Pytorch implementation of MISR-GRU, a deep neural network for multi image super-resolution (MISR), for ProbaV Super Resolution Competition\n\n- [MSDTGP](https://github.com/XY-boy/MSDTGP) -\u003e Satellite Video Super-Resolution via Multiscale Deformable Convolution Alignment and Temporal Grouping Projection\n\n- [proba-v-super-resolution-challenge](https://github.com/cedricoeldorf/proba-v-super-resolution-challenge) -\u003e Solution to ESA's satellite imagery super resolution challenge\n\n- [PROBA-V-Super-Resolution](https://github.com/spicy-mama/PROBA-V-Super-Resolution) -\u003e solution using a custom deep learning architecture\n\n- [satlas-super-resolution](https://github.com/allenai/satlas-super-resolution) -\u003e Satlas Super Resolution: model is an adaptation of ESRGAN, with changes that allow the input to be a time series of Sentinel-2 images.\n\n- [MISR Remote Sensing SRGAN](https://github.com/simon-donike/Remote-Sensing-SRGAN) -\u003e PyTorch SRGAN for RGB Remote Sensing imagery, performing both SISR and MISR. MISR implementation inspired by RecursiveNet (HighResNet). Includes pretrained Checkpoints.\n\n- [MISR-S2](https://github.com/aimiokab/MISR-S2) -\u003e Cross-sensor super-resolution of irregularly sampled Sentinel-2 time series\n\n### Single image super-resolution (SISR)\n\n- [Swin2-MoSE](https://github.com/IMPLabUniPr/swin2-mose) -\u003e Swin2-MoSE: A New Single Image Super-Resolution Model for Remote Sensing\n\n- [sentinel2_superresolution](https://github.com/Evoland-Land-Monitoring-Evolution/sentinel2_superresolution) -\u003e Super-resolution of 10 Sentinel-2 bands to 5-meter resolution, starting from L1C or L2A (Theia format) products. Trained on Sen2Venµs\n\n- [Super Resolution for Satellite Imagery - srcnn repo](https://github.com/WarrenGreen/srcnn)\n\n- [TensorFlow implementation of \"Accurate Image Super-Resolution Using Very Deep Convolutional Networks\" adapted for working with geospatial data](https://github.com/CosmiQ/VDSR4Geo)\n\n- [Random Forest Super-Resolution (RFSR repo)](https://github.com/jshermeyer/RFSR) including [sample data](https://github.com/jshermeyer/RFSR/tree/master/SampleImagery)\n\n- [Enhancing Sentinel 2 images by combining Deep Image Prior and Decrappify](https://medium.com/omdena/pushing-the-limits-of-open-source-data-enhancing-satellite-imagery-through-deep-learning-9d8a3bbc0e0a). Repo for [deep-image-prior](https://github.com/DmitryUlyanov/deep-image-prior) and article on [decrappify](https://www.fast.ai/2019/05/03/decrappify/)\n\n- [Image Super-Resolution using an Efficient Sub-Pixel CNN](https://keras.io/examples/vision/super_resolution_sub_pixel/) -\u003e the keras docs have a great tutorial on this light weight but well performing model\n\n- [super-resolution-using-gan](https://github.com/saraivaufc/super-resolution-using-gan) -\u003e Super-Resolution of Sentinel-2 Using Generative Adversarial Networks\n\n- [Super-resolution of Multispectral Satellite Images Using Convolutional Neural Networks](https://up42.com/blog/tech/super-resolution-of-multispectral-satellite-images-using-convolutional)\n\n- [Multi-temporal Super-Resolution on Sentinel-2 Imagery](https://medium.com/sentinel-hub/multi-temporal-super-resolution-on-sentinel-2-imagery-6089c2b39ebc) using HighRes-Net, [repo](https://github.com/sentinel-hub/multi-temporal-super-resolution)\n\n- [SSPSR-Pytorch](https://github.com/junjun-jiang/SSPSR) -\u003e A spatial-spectral prior deep network for single hyperspectral image super-resolution\n\n- [Sentinel-2 Super-Resolution: High Resolution For All (Bands)](https://up42.com/blog/tech/sentinel-2-superresolution)\n\n- [CinCGAN](https://github.com/Junshk/CinCGAN-pytorch) -\u003e Unsupervised Image Super-Resolution using Cycle-in-Cycle Generative Adversarial Networks\n\n- [Satellite-image-SRGAN using PyTorch](https://github.com/xjohnxjohn/Satellite-image-SRGAN)\n\n- [EEGAN](https://github.com/kuijiang0802/EEGAN) -\u003e Edge Enhanced GAN For Remote Sensing Image Super-Resolution, TensorFlow 1.1\n\n- [PECNN](https://github.com/kuijiang0802/PECNN) -\u003e A Progressively Enhanced Network for Video Satellite Imagery Super-Resolution, minimal documentation\n\n- [hs-sr-tvtv](https://github.com/marijavella/hs-sr-tvtv) -\u003e Enhanced Hyperspectral Image Super-Resolution via RGB Fusion and TV-TV Minimization\n\n- [sr4rs](https://github.com/remicres/sr4rs) -\u003e Super resolution for remote sensing, with pre-trained model for Sentinel-2, SRGAN-inspired\n\n- [RFSR_TGRS](https://github.com/wxywhu/RFSR_TGRS) -\u003e Hyperspectral Image Super-Resolution via Recurrent Feedback Embedding and Spatial-Spectral Consistency Regularization\n\n- [SEN2VENµS](https://zenodo.org/record/6514159#.YoRxM5PMK3I) -\u003e a dataset for the training of Sentinel-2 super-resolution algorithms. With [paper](https://www.mdpi.com/2306-5729/7/7/96)\n\n- [TransENet](https://github.com/Shaosifan/TransENet) -\u003e Transformer-based Multi-Stage Enhancement for Remote Sensing Image Super-Resolution\n\n - [SG-FBGAN](https://github.com/hanlinwu/SG-FBGAN) -\u003e Remote Sensing Image Super-Resolution via Saliency-Guided Feedback GANs\n\n- [finetune_ESRGAN](https://github.com/johnjaniczek/finetune_ESRGAN) -\u003e finetune the ESRGAN super resolution generator for remote sensing images and video\n\n- [MIP](https://github.com/jiaming-wang/MIP) -\u003e Unsupervised Remote Sensing Super-Resolution via Migration Image Prior\n\n- [Optical-RemoteSensing-Image-Resolution](https://github.com/wenjiaXu/Optical-RemoteSensing-Image-Resolution) -\u003e Deep Memory Connected Neural Network for Optical Remote Sensing Image Restoration. Two applications: Gaussian image denoising and single image super-resolution\n\n- [HSENet](https://github.com/Shaosifan/HSENet) -\u003e Hybrid-Scale Self-Similarity Exploitation for Remote Sensing Image Super-Resolution\n\n- [SR_RemoteSensing](https://github.com/Jing25/SR_RemoteSensing) -\u003e Super-Resolution deep learning models for remote sensing data based on [BasicSR](https://github.com/XPixelGroup/BasicSR)\n\n- [RSI-Net](https://github.com/EricBrock/RSI-Net) -\u003e A Deep Multi-task Convolutional Neural Network for Remote Sensing Image Super-resolution and Colorization\n\n- [EDSR-Super-Resolution](https://github.com/RakeshRaj97/EDSR-Super-Resolution) -\u003e EDSR model using PyTorch applied to satellite imagery\n\n- [CycleCNN](https://github.com/haopzhang/CycleCNN) -\u003e Nonpairwise-Trained Cycle Convolutional Neural Network for Single Remote Sensing Image Super-Resolution\n\n- [SISR with with Real-World Degradation Modeling](https://github.com/zhangjizhou-bit/Single-image-Super-Resolution-of-Remote-Sensing-Images-with-Real-World-Degradation-Modeling) -\u003e Single-Image Super Resolution of Remote Sensing Images with Real-World Degradation Modeling\n\n- [pixel-smasher](https://github.com/ekcomputer/pixel-smasher) -\u003e Super-Resolution Surface Water Mapping on the Canadian Shield Using Planet CubeSat Images and a Generative Adversarial Network\n\n- [satellite-image-super-resolution](https://github.com/farahmand-m/satellite-image-super-resolution) -\u003e A Comparative Study on CNN-Based Single-Image Super-Resolution Techniques for Satellite Images\n\n- [SatelliteSR](https://github.com/kmalhan/SatelliteSR) -\u003e comparison of a number of techniques on the DOTA dataset\n\n- [Image-Super-Resolution](https://github.com/Elangoraj/Image-Super-Resolution) -\u003e Super resolution RESNET network\n\n- [Unsupervised Super Resolution for Sentinel-2 satellite imagery](https://github.com/savassif/Thesis) -\u003e using Deep Image Prior (DIP), Zero-Shot Super Resolution (ΖSSR) \u0026 Degradation-Aware Super Resolution (DASR)\n\n- [Spectral Super-Resolution of Satellite Imagery with Generative Adversarial Networks](https://github.com/ImDanielRojas/thesis)\n\n- [Super resolution using GAN / 4x Improvement](https://github.com/purijs/satellite-superresolution) -\u003e applied to Sentinel 2\n\n- [rs-esrgan](https://github.com/luissalgueiro/rs-esrgan) -\u003e RS-ESRGAN: Super-Resolution of Sentinel-2 Imagery Using Generative Adversarial Networks\n\n- [TS-RSGAN](https://github.com/yicrane/TS-RSGAN) -\u003e Super-Resolution of Remote Sensing Images for ×4 Resolution without Reference Images. Applied to Sentinel-2\n\n- [CDCR](https://github.com/Suanmd/CDCR) -\u003e Combining Discrete and Continuous Representation: Scale-Arbitrary Super-Resolution for Satellite Images\n\n- [FunSR](https://github.com/KyanChen/FunSR) -\u003e cContinuous Remote Sensing Image Super-Resolution based on Context Interaction in Implicit Function Space\n\n- [HAUNet_RSISR](https://github.com/likakakaka/HAUNet_RSISR) -\u003e Hybrid Attention-Based U-Shaped Network for Remote Sensing Image Super-Resolution\n\n- [L1BSR](https://github.com/centreborelli/L1BSR) -\u003e Exploiting Detector Overlap for Self-Supervised SISR of Sentinel-2 L1B Imagery\n\n- [Deep-Harmonization](https://github.com/venkatesh-thiru/Deep-Harmonization) -\u003e Deep Learning-based Harmonization and Super-Resolution of Landsat-8 and Sentinel-2 images\n\n- [SGDM](https://github.com/wwangcece/SGDM) -\u003e Semantic Guided Large Scale Factor Remote Sensing Image Super-resolution with Generative Diffusion Prior\n\n- [SEN2SR](https://github.com/ESAOpenSR/SEN2SR) -\u003e A Python package to super-resolve Sentinel-2 satellite imagery up to 2.5 meters.\n  \n### Super-resolution - Miscellaneous\n\n- [The value of super resolution — real world use case](https://medium.com/sentinel-hub/the-value-of-super-resolution-real-world-use-case-2ba811f4cd7f) -\u003e Medium article on parcel boundary detection with super-resolved satellite imagery\n\n- [Super-Resolution on Satellite Imagery using Deep Learning](https://medium.com/the-downlinq/super-resolution-on-satellite-imagery-using-deep-learning-part-1-ec5c5cd3cd2) -\u003e Nov 2016 blog post by CosmiQ Works with a nice introduction to the topic. Proposes and demonstrates a new architecture with perturbation layers with practical guidance on the methodology and [code](https://github.com/CosmiQ/super-resolution). [Three part series](https://medium.com/the-downlinq/super-resolution-on-satellite-imagery-using-deep-learning-part-3-2e2f61eee1d3)\n\n- [Introduction to spatial resolution](https://medium.com/sentinel-hub/the-most-misunderstood-words-in-earth-observation-d0106adbe4b0)\n\n- [Awesome-Super-Resolution](https://github.com/ptkin/Awesome-Super-Resolution) -\u003e another 'awesome' repo, getting a little out of date now\n\n- [Super-Resolution (python) Utilities for managing large satellite images](https://github.com/jshermeyer/SR_Utils)\n\n- [pytorch-enhance](https://github.com/isaaccorley/pytorch-enhance) -\u003e Library of Image Super-Resolution Models, Datasets, and Metrics for Benchmarking or Pretrained Use. Also [checkout this implementation in Jax](https://github.com/isaaccorley/jax-enhance)\n\n- [Super Resolution in OpenCV](https://learnopencv.com/super-resolution-in-opencv/)\n\n- [AI-based Super resolution and change detection to enforce Sentinel-2 systematic usage](https://medium.com/@sistema_gmbh/ai-based-super-resolution-and-change-detection-to-enforce-sentinel-2-systematic-usage-65aa37d0365) -\u003e Worldview-2 images (2m) were used to create a reference dataset and increase the spatial resolution of the Copernicus sensor from 10m to 5m\n\n- [SRCDNet](https://github.com/liumency/SRCDNet) -\u003e Super-resolution-based Change Detection Network with Stacked Attention Module for Images with Different Resolutions. SRCDNet is designed to learn and predict change maps from bi-temporal images with different resolutions\n\n- [Model-Guided Deep Hyperspectral Image Super-resolution](https://github.com/chengerr/Model-Guided-Deep-Hyperspectral-Image-Super-resolution) -\u003e code accompanying the paper: Model-Guided Deep Hyperspectral Image Super-Resolution\n\n- [Super-resolving beyond satellite hardware](https://github.com/smpetrie/superres) -\u003e [paper](https://arxiv.org/abs/2103.06270) assessing SR performance in reconstructing realistically degraded satellite images\n\n- [satellite-pixel-synthesis-pytorch](https://github.com/KellyYutongHe/satellite-pixel-synthesis-pytorch) -\u003e PyTorch implementation of NeurIPS 2021 paper: Spatial-Temporal Super-Resolution of Satellite Imagery via Conditional Pixel Synthesis\n\n- [SRE-HAN](https://github.com/bostankhan6/SRE-HAN) -\u003e Squeeze-and-Residual-Excitation Holistic Attention Network improves super-resolution (SR) on remote-sensing imagery compared to other state-of-the-art attention-based SR models\n\n- [satsr](https://github.com/deephdc/satsr) -\u003e A project to perform super-resolution on multispectral images from any satellite, including Sentinel 2, Landsat 8, VIIRS \u0026MODIS\n\n- [OLI2MSI](https://github.com/wjwjww/OLI2MSI) -\u003e dataset for remote sensing imagery super-resolution composed of Landsat8-OLI and Sentinel2-MSI images\n\n- [MMSR](https://github.com/palmdong/MMSR) -\u003e Learning Mutual Modulation for Self-Supervised Cross-Modal Super-Resolution\n\n- [HSRnet](https://github.com/liangjiandeng/HSRnet) -\u003e Hyperspectral Image Super-resolution via Deep Spatio-spectral Attention Convolutional Neural Networks\n\n- [RRSGAN](https://github.com/dongrunmin/RRSGAN) -\u003e RRSGAN: Reference-Based Super-Resolution for Remote Sensing Image\n\n- [HDR-DSP-SR](https://github.com/centreborelli/HDR-DSP-SR) -\u003e Self-supervised multi-image super-resolution for push-frame satellite images\n\n- [GAN-HSI-SR](https://github.com/ZhuangChen25674/GAN-HSI-SR) -\u003e Hyperspectral Image Super-Resolution by Band Attention Through Adversarial Learning\n\n- [SalDRN](https://github.com/hanlinwu/SalDRN) -\u003e Lightweight Stepless Super-Resolution of Remote Sensing Images via Saliency-Aware Dynamic Routing Strategy [paper](https://arxiv.org/abs/2210.07598)\n\n- [BlindSRSNF](https://github.com/hanlinwu/BlindSRSNF) -\u003e Conditional Stochastic Normalizing Flows for Blind Super-Resolution of Remote Sensing Images [paper](https://arxiv.org/abs/2210.07751)\n\n#\n## Pansharpening\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/pansharpen.png\" width=\"500\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003ePansharpening example with a resolution difference of factor 4.\u003c/b\u003e\n\u003c/p\u003e\n\nPansharpening is a data fusion method that merges the high spatial detail from a high-resolution panchromatic image with the rich spectral information from a lower-resolution multispectral image. The result is a single, high-resolution color image that retains both the sharpness of the panchromatic band and the color information of the multispectral bands. This process enhances the spatial resolution while preserving the spectral qualities of the original images. [Image source](https://www.researchgate.net/publication/308121983_Computer_Vision_for_Large_Format_Digital_Aerial_Cameras)\n\n- Several algorithms described [in the ArcGIS docs](http://desktop.arcgis.com/en/arcmap/10.3/manage-data/raster-and-images/fundamentals-of-panchromatic-sharpening.htm), with the simplest being taking the mean of the pan and RGB pixel value.\n\n- [PGCU](https://github.com/Zeyu-Zhu/PGCU) -\u003e Probability-based Global Cross-modal Upsampling for Pansharpening\n\n- [rio-pansharpen](https://github.com/mapbox/rio-pansharpen) -\u003e pansharpening Landsat scenes\n\n- [Simple-Pansharpening-Algorithms](https://github.com/ThomasWangWeiHong/Simple-Pansharpening-Algorithms)\n\n- [Working-For-Pansharpening](https://github.com/yuanmaoxun/Working-For-Pansharpening) -\u003e long list of pansharpening methods and update of [Awesome-Pansharpening](https://github.com/Lihui-Chen/Awesome-Pansharpening)\n\n- [PSGAN](https://github.com/liuqingjie/PSGAN) -\u003e A Generative Adversarial Network for Remote Sensing Image Pan-sharpening\n\n- [Pansharpening-by-Convolutional-Neural-Network](https://github.com/ThomasWangWeiHong/Pansharpening-by-Convolutional-Neural-Network)\n\n- [PBR_filter](https://github.com/dbuscombe-usgs/PBR_filter) -\u003e Pansharpening by Background Removal algorithm for sharpening RGB images\n\n- [py_pansharpening](https://github.com/codegaj/py_pansharpening) -\u003e multiple algorithms implemented in python\n\n- [Deep-Learning-PanSharpening](https://github.com/xyc19970716/Deep-Learning-PanSharpening) -\u003e deep-learning based pan-sharpening code package, we reimplemented include PNN, MSDCNN, PanNet, TFNet, SRPPNN, and our purposed network DIPNet\n\n- [HyperTransformer](https://github.com/wgcban/HyperTransformer) -\u003e A Textural and Spectral Feature Fusion Transformer for Pansharpening\n\n- [DIP-HyperKite](https://github.com/wgcban/DIP-HyperKite) -\u003e Hyperspectral Pansharpening Based on Improved Deep Image Prior and Residual Reconstruction\n\n- [D2TNet](https://github.com/Meiqi-Gong/D2TNet) -\u003e A ConvLSTM Network with Dual-direction Transfer for Pan-sharpening\n\n- [PanColorGAN-VHR-Satellite-Images](https://github.com/esertel/PanColorGAN-VHR-Satellite-Images) -\u003e Rethinking CNN-Based Pansharpening: Guided Colorization of Panchromatic Images via GANs\n\n- [MTL_PAN_SEG](https://github.com/andrewekhalel/MTL_PAN_SEG) -\u003e Multi-task deep learning for satellite image pansharpening and segmentation\n\n- [Z-PNN](https://github.com/matciotola/Z-PNN) -\u003e Pansharpening by convolutional neural networks in the full resolution framework\n\n- [GTP-PNet](https://github.com/HaoZhang1018/GTP-PNet) -\u003e GTP-PNet: A residual learning network based on gradient transformation prior for pansharpening\n\n- [UDL](https://github.com/XiaoXiao-Woo/UDL) -\u003e Dynamic Cross Feature Fusion for Remote Sensing Pansharpening\n\n- [PSData](https://github.com/yisun98/PSData) -\u003e A Large-Scale General Pan-sharpening DataSet, which contains PSData3 (QB, GF-2, WV-3) and PSData4 (QB, GF-1, GF-2, WV-2).\n\n- [AFPN](https://github.com/yisun98/AFPN) -\u003e Adaptive Detail Injection-Based Feature Pyramid Network For Pan-sharpening\n\n- [pan-sharpening](https://github.com/yisun98/pan-sharpening) -\u003e multiple methods demonstrated for multispectral and panchromatic images\n\n- [PSGan-Family](https://github.com/zhysora/PSGan-Family) -\u003e PSGAN: A Generative Adversarial Network for Remote Sensing Image Pan-Sharpening\n\n- [PanNet-Landsat](https://github.com/oyam/PanNet-Landsat) -\u003e A Deep Network Architecture for Pan-Sharpening\n\n- [DLPan-Toolbox](https://github.com/liangjiandeng/DLPan-Toolbox) -\u003e  Machine Learning in Pansharpening: A Benchmark, from Shallow to Deep Networks\n\n- [LPPN](https://github.com/ChengJin-git/LPPN) -\u003e Laplacian pyramid networks: A new approach for multispectral pansharpening\n\n- [S2_SSC_CNN](https://github.com/hvn2/S2_SSC_CNN) -\u003e Zero-shot Sentinel-2 Sharpening Using A Symmetric Skipped Connection Convolutional Neural Network\n\n- [S2S_UCNN](https://github.com/hvn2/S2S_UCNN) -\u003e Sentinel 2 sharpening using a single unsupervised convolutional neural network with MTF-Based degradation model\n\n- [SSE-Net](https://github.com/RSMagneto/SSE-Net) -\u003e Spatial and Spectral Extraction Network With Adaptive Feature Fusion for Pansharpening\n\n- [UCGAN](https://github.com/zhysora/UCGAN) -\u003e Unsupervised Cycle-consistent Generative Adversarial Networks for Pan-sharpening\n\n- [GCPNet](https://github.com/Keyu-Yan/GCPNet) -\u003e When Pansharpening Meets Graph Convolution Network and Knowledge Distillation\n\n- [PanFormer](https://github.com/zhysora/PanFormer) -\u003e PanFormer: a Transformer Based Model for Pan-sharpening\n\n- [Pansharpening](https://github.com/nithin-gr/Pansharpening) -\u003e Pansformers: Transformer-Based Self-Attention Network for Pansharpening\n\n- [Sentinel-2 Band Pan-Sharpening](https://github.com/purijs/Sentinel-2-Superresolution)\n\n- [UAPN](https://github.com/keviner1/UAPN) -\u003e Official PyTorch implementation of our TGRS paper: Deep Adaptive Pansharpening via Uncertainty-aware Image Fusion.[Paper link](https://ieeexplore.ieee.org/document/10106462)\n\n#\n## Image-to-image translation\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/translation.png\" width=\"500\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003e(left) Sentinel-1 SAR input, (middle) translated to RGB and (right) Sentinel-2 true RGB image for comparison.\u003c/b\u003e\n\u003c/p\u003e\n\nImage-to-image translation is a crucial aspect of computer vision that utilizes machine learning models to transform an input image into a new, distinct output image. In the field of remote sensing, it plays a significant role in bridging the gap between different imaging domains, such as converting Synthetic Aperture Radar (SAR) images into RGB (Red Green Blue) images. This technology has a wide range of applications, including improving image quality, filling in missing information, and facilitating cross-domain image analysis and comparison. By leveraging deep learning algorithms, image-to-image translation has become a powerful tool in the arsenal of remote sensing researchers and practitioners. [Image source](https://www.researchgate.net/publication/335648375_SAR-to-Optical_Image_Translation_Using_Supervised_Cycle-Consistent_Adversarial_Networks)\n\n- [How to Develop a Pix2Pix GAN for Image-to-Image Translation](https://machinelearningmastery.com/how-to-develop-a-pix2pix-gan-for-image-to-image-translation/) -\u003e how to develop a Pix2Pix model for translating satellite photographs to Google map images. A good intro to GANS\n\n- [A growing problem of ‘deepfake geography’: How AI falsifies satellite images](https://www.washington.edu/news/2021/04/21/a-growing-problem-of-deepfake-geography-how-ai-falsifies-satellite-images/)\n\n- [Kaggle Pix2Pix Maps](https://www.kaggle.com/datasets/alincijov/pix2pix-maps) -\u003e dataset for pix2pix to take a google map satellite photo and build a street map\n\n- [guided-deep-decoder](https://github.com/tuezato/guided-deep-decoder) -\u003e With guided deep decoder, you can solve different image pair fusion problems, allowing super-resolution, pansharpening or denoising\n\n- [hackathon-ci-2020](https://github.com/paulaharder/hackathon-ci-2020) -\u003e generate nighttime imagery from infrared observations\n\n- [satellite-to-satellite-translation](https://github.com/anonymous-ai-for-earth/satellite-to-satellite-translation) -\u003e VAE-GAN architecture for unsupervised image-to-image translation with shared spectral reconstruction loss. Model is trained on GOES-16/17 and Himawari-8 L1B data\n\n- [Pytorch implementation of UNet for converting aerial satellite images into google maps kinda images](https://github.com/greed2411/unet_pytorch)\n\n- [Seamless-Satellite-image-Synthesis](https://github.com/Misaliet/Seamless-Satellite-image-Synthesis) -\u003e generate abitrarily large RGB images from a map\n\n- [How to Develop a Pix2Pix GAN for Image-to-Image Translation](https://machinelearningmastery.com/how-to-develop-a-pix2pix-gan-for-image-to-image-translation/) -\u003e article on machinelearningmastery.com\n\n- [Satellite-Imagery-to-Map-Translation-using-Pix2Pix-GAN-framework](https://github.com/anh-nn01/Satellite-Imagery-to-Map-Translation-using-Pix2Pix-GAN-framework)\n\n- [RSIT_SRM_ISD](https://github.com/summitgao/RSIT_SRM_ISD) -\u003e PyTorch implementation of Remote sensing image translation via style-based recalibration module and improved style discriminator\n\n- [pix2pix_google_maps](https://github.com/manishemirani/pix2pix_google_maps) -\u003e Converts satellite images to map images using pix2pix models\n\n- [sar2color-igarss2018-chainer](https://github.com/enomotokenji/sar2color-igarss2018-chainer) -\u003e Image Translation Between Sar and Optical Imagery with Generative Adversarial Nets\n\n- [HSI2RGB](https://github.com/JakobSig/HSI2RGB) -\u003e Create realistic looking RGB images using remote sensing hyperspectral images\n\n- [sat_to_map](https://github.com/shagunuppal/sat_to_map) -\u003e Learning mappings to generate city maps images from corresponding satellite images\n\n- [pix2pix-GANs](https://github.com/shashi7679/pix2pix-GANs) -\u003e Generate Map using Satellite Image \u0026 PyTorch\n\n- [map-sat](https://github.com/miquel-espinosa/map-sat) -\u003e Generate Your Own Scotland: Satellite Image Generation Conditioned on Maps\n\n#\n## Data fusion\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/fusion.png\" width=\"800\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eIllustration of a fusion workflow.\u003c/b\u003e\n\u003c/p\u003e\n\nData fusion is a technique for combining information from different sources such as Synthetic Aperture Radar (SAR), optical imagery, and non-imagery data such as Internet of Things (IoT) sensor data. The integration of diverse data sources enables data fusion to overcome the limitations of individual sources, leading to the creation of models that are more accurate and informative than those constructed from a single source. [Image source](https://www.mdpi.com/2072-4292/14/18/4458)\n\n- [Awesome-Data-Fusion-for-Remote-Sensing](https://github.com/px39n/Awesome-Data-Fusion-for-Remote-Sensing)\n\n- [UDALN_GRSL](https://github.com/JiaxinLiCAS/UDALN_GRSL) -\u003e Deep Unsupervised Blind Hyperspectral and Multispectral Data Fusion\n\n- [CropTypeMapping](https://github.com/ellaampy/CropTypeMapping) -\u003e Crop type mapping from optical and radar (Sentinel-1\u00262) time series using attention-based deep learning\n\n- [Multimodal-Remote-Sensing-Toolkit](https://github.com/likyoo/Multimodal-Remote-Sensing-Toolkit) -\u003e uses Hyperspectral and LiDAR Data\n\n- [Aerial-Template-Matching](https://github.com/m-hamza-mughal/Aerial-Template-Matching) -\u003e development of an algorithm for template Matching on aerial imagery applied to UAV dataset\n\n- [DS_UNet](https://github.com/SebastianHafner/DS_UNet) -\u003e Sentinel-1 and Sentinel-2 Data Fusion for Urban Change Detection using a Dual Stream U-Net, uses Onera Satellite Change Detection dataset\n\n- [DDA_UrbanExtraction](https://github.com/SebastianHafner/DDA_UrbanExtraction) -\u003e Unsupervised Domain Adaptation for Global Urban Extraction using Sentinel-1 and Sentinel-2 Data\n\n- [swinstfm](https://github.com/LouisChen0104/swinstfm) -\u003e Remote Sensing Spatiotemporal Fusion using Swin Transformer\n\n- [LoveCS](https://github.com/Junjue-Wang/LoveCS) -\u003e Cross-sensor domain adaptation for high-spatial resolution urban land-cover mapping: from airborne to spaceborne imagery\n\n- [comingdowntoearth](https://github.com/aysim/comingdowntoearth) -\u003e Implementation of 'Coming Down to Earth: Satellite-to-Street View Synthesis for Geo-Localization'\n\n- [Matching between acoustic and satellite images](https://github.com/giovgiac/neptune)\n\n- [MapRepair](https://github.com/zorzi-s/MapRepair) -\u003e Deep Cadastre Maps Alignment and Temporal Inconsistencies Fix in Satellite Images\n\n- [Compressive-Sensing-and-Deep-Learning-Framework](https://github.com/rahulgite94/Compressive-Sensing-and-Deep-Learning-Framework) -\u003e  Compressive Sensing is used as an initial guess to combine data from multiple sources, with LSTM used to refine the result\n\n- [DeepSim](https://github.com/wangxiaodiu/DeepSim) -\u003e DeepSIM: GPS Spoofing Detection on UAVs using Satellite Imagery Matching\n\n- [MHF-net](https://github.com/XieQi2015/MHF-net) -\u003e Multispectral and Hyperspectral Image Fusion by MS/HS Fusion Net\n\n- [Remote_Sensing_Image_Fusion](https://github.com/huangshanshan33/Remote_Sensing_Image_Fusion) -\u003e Semi-Supervised Remote Sensing Image Fusion Using Multi-Scale Conditional Generative Adversarial network with Siamese Structure\n\n- [CNNs for Multi-Source Remote Sensing Data Fusion](https://github.com/yyyyangyi/CNNs-for-Multi-Source-Remote-Sensing-Data-Fusion) -\u003e Single-stream CNN with Learnable Architecture for Multi-source Remote Sensing Data\n\n- [Deep Generative Reflectance Fusion](https://github.com/Cervest/ds-generative-reflectance-fusion) -\u003e Achieving Landsat-like reflectance at any date by fusing Landsat and MODIS surface reflectance with deep generative models\n\n- [IEEE_TGRS_MDL-RS](https://github.com/danfenghong/IEEE_TGRS_MDL-RS) -\u003e  More Diverse Means Better: Multimodal Deep Learning Meets Remote-Sensing Imagery Classification\n\n- [SSRNET](https://github.com/hw2hwei/SSRNET) -\u003e SSR-NET: Spatial-Spectral Reconstruction Network for Hyperspectral and Multispectral Image Fusion\n\n- [cross-view-image-matching](https://github.com/kregmi/cross-view-image-matching) -\u003e Bridging the Domain Gap for Ground-to-Aerial Image Matching\n\n- [CoF-MSMG-PCNN](https://github.com/WeiTan1992/CoF-MSMG-PCNN) -\u003e Remote Sensing Image Fusion via Boundary Measured Dual-Channel PCNN in Multi-Scale Morphological Gradient Domain\n\n- [robust_matching_network_on_remote_sensing_imagery_pytorch](https://github.com/mrk1992/robust_matching_network_on_remote_sensing_imagery_pytorch) -\u003e A Robust Matching Network for Gradually Estimating Geometric Transformation on Remote Sensing Imagery\n\n- [edcstfn](https://github.com/theonegis/edcstfn) -\u003e An Enhanced Deep Convolutional Model for Spatiotemporal Image Fusion\n\n- [ganstfm](https://github.com/theonegis/ganstfm) -\u003e A Flexible Reference-Insensitive Spatiotemporal Fusion Model for Remote Sensing Images Using Conditional Generative Adversarial Network\n\n- [CMAFF](https://github.com/DocF/CMAFF) -\u003e Cross-Modality Attentive Feature Fusion for Object Detection in Multispectral Remote Sensing Imagery\n\n- [SOLC](https://github.com/yisun98/SOLC) -\u003e MCANet: A joint semantic segmentation framework of optical and SAR images for land use classification. Uses [WHU-OPT-SAR-dataset](https://github.com/AmberHen/WHU-OPT-SAR-dataset)\n\n- [MFT](https://github.com/AnkurDeria/MFT) -\u003e Multimodal Fusion Transformer for Remote Sensing Image Classification\n\n- [ISPRS_S2FL](https://github.com/danfenghong/ISPRS_S2FL) -\u003e Multimodal Remote Sensing Benchmark Datasets for Land Cover Classification with A Shared and Specific Feature Learning Model\n\n- [HSHT-Satellite-Imagery-Synthesis](https://github.com/yuvalofek/HSHT-Satellite-Imagery-Synthesis) -\u003e Improving Flood Maps by Increasing the Temporal Resolution of Satellites Using Hybrid Sensor Fusion\n\n- [MDC](https://github.com/Kasra2020/MDC) -\u003e Unsupervised Data Fusion With Deeper Perspective: A Novel Multisensor Deep Clustering Algorithm\n\n- [FusAtNet](https://github.com/ShivamP1993/FusAtNet) -\u003e FusAtNet: Dual Attention based SpectroSpatial Multimodal Fusion Network for Hyperspectral and LiDAR Classification\n\n- [AMM-FuseNet](https://github.com/oktaykarakus/ReSIF/tree/main/AMM-FuseNet) -\u003e Attention-Based Multi-Modal Image Fusion Network for Land Cover Mapping\n\n- [MANet](https://github.com/caohuimin/MANet) -\u003e MANet: A Network Architecture for Remote Sensing Spatiotemporal Fusion Based on Multiscale and Attention Mechanisms\n\n- [DCSA-Net](https://github.com/Julia90/DCSA-Net) -\u003e Dynamic Convolution Self-Attention Network for Land-Cover Classification in VHR Remote-Sensing Images\n\n- [deforestation-from-data-fusion](https://github.com/felferrari/deforestation-from-data-fusion) -\u003e Fusing Sentinel-1 and Sentinel-2 images for deforestation detection in the Brazilian Amazon under diverse cloud conditions\n\n- [sct-fusion](https://git.tu-berlin.de/rsim/sct-fusion) -\u003e Transformer-based Multi-Modal Learning for Multi Label Remote Sensing Image Classification\n\n- [RSI-MMSegmentation](https://github.com/EarthNets/RSI-MMSegmentation) -\u003e GAMUS: A Geometry-aware Multi-modal Semantic Segmentation Benchmark for Remote Sensing Data\n\n- [dfc2022-baseline](https://github.com/isaaccorley/dfc2022-baseline) -\u003e baseline solution to the 2022 IEEE GRSS Data Fusion Contest (DFC2022) using TorchGeo, PyTorch Lightning, and Segmentation Models PyTorch to train a U-Net with a ResNet-18 backbone and a loss function of Focal + Dice loss to perform semantic segmentation on the DFC2022 dataset\n\n- [multiviewRS-models](https://github.com/fmenat/multiviewRS-models) -\u003e List of multi-view fusion learning models proposed for remote sensing (RS) multi-view data\n\n#\n## Generative networks\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/gan.png\" width=\"500\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eExample generated images using a GAN.\u003c/b\u003e\n\u003c/p\u003e\n\nGenerative networks (e.g. GANs) aim to generate new, synthetic data that appears similar to real-world data. This generated data can be used for a wide range of purposes, including data augmentation, data imbalance correction, and filling in missing or corrupted data. Including generating synthetic data can improve the performance of remote sensing algorithms and models, leading to more accurate and reliable results. [Image source](https://arxiv.org/abs/2207.14580)\n\n- [Using Generative Adversarial Networks to Address Scarcity of Geospatial Training Data](https://medium.com/radiant-earth-insights/using-generative-adversarial-networks-to-address-scarcity-of-geospatial-training-data-e61cacec986e) -\u003e GAN perform better than CNN in segmenting land cover classes outside of the training dataset (article, no code)\n\n- [Building-A-Nets](https://github.com/lixiang-ucas/Building-A-Nets) -\u003e robust building extraction from high-resolution remote sensing images with adversarial networks\n\n- [GANmapper](https://github.com/ualsg/GANmapper) -\u003e a building footprint generator using Generative Adversarial Networks\n\n- [CSA-CDGAN](https://github.com/wangle53/CSA-CDGAN) -\u003e Channel Self-Attention Based Generative Adversarial Network for Change Detection of Remote Sensing Images\n\n- [DSGAN](https://github.com/lzhengchun/DSGAN) -\u003e a conditinal GAN for dynamic precipitation downscaling\n\n- [MarsGAN](https://github.com/kheyer/MarsGAN) -\u003e GAN trained on satellite photos of Mars\n\n- [HC_ADGAN](https://github.com/summitgao/HC_ADGAN) -\u003e codes for the paper Adaptive Dropblock Enhanced GenerativeAdversarial Networks for Hyperspectral Image Classification\n\n- [SCALAE](https://github.com/LendelTheGreat/SCALAE) -\u003e Formatting the Landscape: Spatial conditional GAN for varying population in satellite imagery. Method to generate satellite imagery from custom 2D population maps\n\n- [Satellite-Image-Forgery-Detection-and-Localization](https://github.com/tailongnguyen/Satellite-Image-Forgery-Detection-and-Localization)\n\n- [STGAN](https://github.com/ermongroup/STGAN) -\u003e PyTorch Implementation of STGAN for Cloud Removal in Satellite Images\n\n- [ds-gan-spatiotemporal-evaluation](https://github.com/Cervest/ds-gan-spatiotemporal-evaluation) -\u003e evaluating use of deep generative models in remote sensing applications\n\n- [GAN-based method to generate high-resolution remote sensing for data augmentation and image classification](https://github.com/weihancug/GAN-based-HRRS-Sample-Generation-for-Image-Classification)\n\n- [Remote-Sensing-Image-Generation](https://github.com/aashishrai3799/Remote-Sensing-Image-Generation) -\u003e Generate RS Images using Generative Adversarial Networks (GAN)\n\n- [RoadDA](https://github.com/LANMNG/RoadDA) -\u003e Stagewise Unsupervised Domain Adaptation with Adversarial Self-Training for Road Segmentation of Remote Sensing Images\n\n- [PSGan-Family](https://github.com/zhysora/PSGan-Family) -\u003e A Generative Adversarial Network for Remote Sensing Image Pan-Sharpening\n\n- [Satellite Image Augmetation with GANs](https://github.com/Oarowolo11/11785-Project) -\u003e Image Augmentation for Satellite Images\n\n- [opt2sar-cyclegan](https://github.com/zzh811/opt2sar-cyclegan) -\u003e Research on SAR image generation method based on non-homologous data\n\n- [sentinel-cgan](https://github.com/softwaremill/sentinel-cgan) -\u003e code for [article](https://blog.softwaremill.com/generative-adversarial-networks-in-satellite-image-datasets-augmentation-b7045d2f51ab): Generative adversarial networks in satellite image datasets augmentation\n\n- [Shoreline_Extraction_GAN](https://github.com/mlundine/Shoreline_Extraction_GAN) -\u003e Shoreline extraction via generative adversarial networks, prediction via LSTMs\n\n- [Landsat8-Sentinel2-Fusion](https://github.com/Rohit18/Landsat8-Sentinel2-Fusion) -\u003e Translating Landsat 8 to Sentinel-2 using a GAN\n\n- [Seg2Sat](https://github.com/RubenGres/Seg2Sat) -\u003e Seg2Sat explores the potential of diffusion algorithms such as StableDiffusion and ControlNet to generate aerial images based on terrain segmentation data\n\n- [SAR2Optical](https://github.com/MuhammedM294/SAR2Optical) -\u003e Transcoding Sentinel-1 SAR to Sentinel-2 using cGAN\n\n- [Urban-Tree-Generator](https://github.com/adnan0819/Urban-Tree-Generator) -\u003e Spatio-Temporal and Generative Deep Learning for Urban Tree Localization and Modeling [paper](https://link.springer.com/article/10.1007/s00371-022-02526-x)\n\n#\n## Autoencoders, dimensionality reduction, image embeddings \u0026 similarity search\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/autoencoder.png\" width=\"600\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eExample of using an autoencoder to create a low dimensional representation of hyperspectral data.\u003c/b\u003e\n\u003c/p\u003e\n\nAutoencoders are a type of neural network that aim to simplify the representation of input data by compressing it into a lower dimensional form. This is achieved through a two-step process of encoding and decoding, where the encoding step compresses the data into a lower dimensional representation, and the decoding step restores the data back to its original form. The goal of this process is to reduce the data's dimensionality, making it easier to store and process, while retaining the essential information. Dimensionality reduction, as the name suggests, refers to the process of reducing the number of dimensions in a dataset. This can be achieved through various techniques such as principal component analysis (PCA) or singular value decomposition (SVD). Autoencoders are one type of neural network that can be used for dimensionality reduction. In the field of computer vision, image embeddings are vector representations of images that capture the most important features of the image. These embeddings can then be used to perform similarity searches, where images are compared based on their features to find similar images. This process can be used in a variety of applications, such as image retrieval, where images are searched based on certain criteria like color, texture, or shape. It can also be used to identify duplicate images in a dataset. [Image source](https://www.mdpi.com/2072-4292/11/7/864)\n\n- [LEt-SNE](https://github.com/meghshukla/LEt-SNE) -\u003e Dimensionality Reduction and visualization technique that compensates for the curse of dimensionality\n\n- [Image-Similarity-Search](https://github.com/spaceml-org/Image-Similarity-Search) -\u003e an app that helps perform super fast image retrieval on PyTorch models for better embedding space interpretability\n\n- [Interactive-TSNE](https://github.com/spaceml-org/Interactive-TSNE) -\u003e a tool that provides a way to visually view a PyTorch model's feature representation for better embedding space interpretability\n\n- [RoofNet](https://github.com/ultysim/RoofNet) -\u003e identify roof age using historical satellite images to lower the customer acquisition cost for new solar installations. Uses a VAE: Variational Autoencoder\n\n- [Visual search over billions of aerial and satellite images](https://arxiv.org/abs/2002.02624)\n\n- [parallax](https://github.com/uber-research/parallax) -\u003e Tool for interactive embeddings visualization\n\n- [Deep-Gapfill](https://github.com/remicres/Deep-Gapfill) -\u003e Official implementation of Optical image gap filling using deep convolutional autoencoder from optical and radar images\n\n- [Mxnet repository for generating embeddings on satellite images](https://github.com/fisch92/Metric-embeddings-for-satellite-image-classification) -\u003e Includes sampling of images, mining algorithms, different architectures, error functions, measures for evaluation.\n\n- [Fine tuning CLIP with Remote Sensing (Satellite) images and captions](https://huggingface.co/blog/fine-tune-clip-rsicd) -\u003e fine tuning CLIP on the [RSICD](https://github.com/201528014227051/RSICD_optimal) image captioning dataset, to enable querying large catalogues in natural language. With [repo](https://github.com/arampacha/CLIP-rsicd), uses 🤗\n\n- [Image search with 🤗 datasets](https://huggingface.co/blog/image-search-datasets) -\u003e tutorial on fine tuning an image search model\n\n- [GRN-SNDL](https://github.com/jiankang1991/GRN-SNDL) -\u003e model the relations between samples (or scenes) by making use of a graph structure which is fed into network learning\n\n- [SauMoCo](https://github.com/jiankang1991/SauMoCo) -\u003e Deep Unsupervised Embedding for Remotely Sensed Images Based on Spatially Augmented Momentum Contrast\n\n- [TGRS_RiDe](https://github.com/jiankang1991/TGRS_RiDe) -\u003e Rotation Invariant Deep Embedding for RemoteSensing Images\n\n- [RaVAEn](https://github.com/spaceml-org/RaVAEn) -\u003e RaVAEn is a lightweight, unsupervised approach for change detection in satellite data based on Variational Auto-Encoders (VAEs) with the specific purpose of on-board deployment\n\n- [Reverse image search using deep discrete feature extraction and locality-sensitive hashing](https://github.com/martenjostmann/deep-discrete-image-retrieval)\n\n- [SNCA_CE](https://github.com/jiankang1991/SNCA_CE) -\u003e Deep Metric Learning based on Scalable Neighborhood Components for Remote Sensing Scene Characterization\n\n- [LandslideDetection-from-satellite-imagery](https://github.com/shulavkarki/LandslideDetection-from-satellite-imagery) -\u003e Using Attention and Autoencoder boosted CNN\n\n- [split-brain-remote-sensing](https://github.com/vladan-stojnic/split-brain-remote-sensing) -\u003e Analysis of Color Space Quantization in Split-Brain Autoencoder for Remote Sensing Image Classification\n\n- [image-similarity-measures](https://github.com/up42/image-similarity-measures) -\u003e Implementation of eight evaluation metrics to access the similarity between two images. [Blog post here](https://up42.com/blog/tech/image-similarity-measures)\n\n- [Large_Scale_GeoVisual_Search](https://github.com/sdhayalk/Large_Scale_GeoVisual_Search) -\u003e ResNet architecture on UC Merced Land Use Dataset with hamming distance for similarity based search\n\n- [geobacter](https://github.com/JakeForsey/geobacter) -\u003e Generates useful feature embeddings for geospatial locations\n\n- [Satellite-Image-Segmentation](https://github.com/kunnalparihar/Satellite-Image-Segmentation) -\u003e the KV-Net model uses this feature of autoencoders to reconnect the disconnected roads\n\n- [Satellite-Image-Enhancement](https://github.com/VNDhanush/Satellite-Image-Enhancement) -\u003e Image enhancement using GAN's and autoencoders\n\n- [Variational-Autoencoder-For-Satellite-Imagery](https://github.com/RayanAAY-ops/Variational-Autoencoder-For-Satellite-Imagery) -\u003e a special VAE to squeeze N images into one single representation with colors segmentating the different objects\n\n- [DINCAE](https://github.com/gher-ulg/DINCAE) -\u003e Data-Interpolating Convolutional Auto-Encoder is a neural network to reconstruct missing data in satellite observations\n\n- [3D_SITS_Clustering](https://github.com/ekalinicheva/3D_SITS_Clustering) -\u003e Unsupervised Satellite Image Time Series Clustering Using Object-Based Approaches and 3D Convolutional Autoencoder\n\n- [sat_cnn](https://github.com/GDSL-UL/sat_cnn) -\u003e Estimating Generalized Measures of Local Neighbourhood Context from Multispectral Satellite Images Using a Convolutional Neural Network. Uses a convolutional autoencoder (CAE)\n\n- [you-are-here](https://github.com/ZhouMengjie/you-are-here) -\u003e You Are Here: Geolocation by Embedding Maps and Images\n\n- [Tensorflow similarity](https://github.com/tensorflow/similarity) -\u003e offers state-of-the-art algorithms for metric learning and all the necessary components to research, train, evaluate, and serve similarity-based models\n\n- [Train SimSiam on Satellite Images](https://docs.lightly.ai/tutorials/package/tutorial_simsiam_esa.html) using lightly.ai to generate embeddings that can be used for data exploration and understanding\n\n- [Airbus_SDC_dup](https://github.com/WillieMaddox/Airbus_SDC_dup) -\u003e Project focused on detecting duplicate regions of overlapping satellite imagery. Applied to Airbus ship detection dataset\n\n- [scale-mae](https://github.com/bair-climate-initiative/scale-mae) -\u003e Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learning\n\n- [Cross-Scale-MAE](https://github.com/aicip/Cross-Scale-MAE) -\u003e code for paper: Cross-Scale MAE: A Tale of Multiscale Exploitation in Remote Sensing\n\n- [satclip](https://github.com/microsoft/satclip) -\u003e A Global, General-Purpose Geographic Location Encoder from Microsoft\n\n- [Astronaut Photography Localization \u0026 Iterative Coregistration](https://earthloc-and-earthmatch.github.io/)\n\n- [rs-cbir](https://github.com/amirafshari/rs-cbir) -\u003e Satellite Image Vector Database and Multimodal Search using fine-tuned ResNet50 on AID dataset\n\n- [TorchSpatial](https://github.com/seai-lab/TorchSpatial) -\u003e A Location Encoding Framework and Benchmark for Spatial Representation Learning\n\n- [experimental-design-multichannel](https://github.com/sbb-gh/experimental-design-multichannel) -\u003e Task-based image channel selection e.g. select most informative hyperspectral wavelengths and perform a task. [Paper](https://openreview.net/forum?id=MloaGA6WwX).\n\n- [PMAA](https://github.com/XavierJiezou/PMAA) -\u003e A Progressive Multi-scale Attention Autoencoder Model for High-Performance Cloud Removal from Multi-temporal Satellite Imagery\n\n- [Temporal MOSAIKS](https://github.com/isaaccorley/temporal-mosaiks) -\u003e Embed2Scale Challenge 4th Place Solution (Training Free!)\n\n- [RANGE](https://github.com/mvrl/RANGE) -\u003e Retrieval Augmented Neural Fields for Multi-Resolution Geo-Embeddings\n\n- [geoclap](https://github.com/mvrl/geoclap) -\u003e Learning Tri-modal Embeddings for Zero-Shot Soundscape Mapping\n\n#\n## Anomaly detection\nAnomaly detection refers to the process of identifying unusual patterns or outliers in satellite or aerial images that do not conform to expected norms. This is crucial in applications such as environmental monitoring, defense surveillance, and urban planning. Machine learning algorithms, particularly unsupervised learning methods, are used to analyze vast amounts of remote sensing data efficiently. These algorithms learn the typical patterns and variations in the data, allowing them to flag anomalies such as unexpected land cover changes, illegal deforestation, or unusual maritime activities. The detection of these anomalies can provide valuable insights for timely decision-making and intervention in various fields.\n\n- [marine-anomaly-detection](https://github.com/lucamarini22/marine-anomaly-detection) -\u003e Semantic segmentation of marine anomalies using semi-supervised learning (FixMatch for semantic segmentation) on Sentinel-2 multispectral images\n\n- [TDD](https://github.com/Jingtao-Li-CVer/TDD) -\u003e One-Step Detection Paradigm for Hyperspectral Anomaly Detection via Spectral Deviation Relationship Learning\n\n- [anomaly-detection-in-SAR-imagery](https://github.com/iamyadavabhishek/anomaly-detection-in-SAR-imagery) -\u003e identify an unknown ship in docks using keras \u0026 retinanet\n\n- [pub-ffi-gan](https://github.com/awweide/pub-ffi-gan) -\u003e Applying generative adversarial networks for anomaly detection in hyperspectral remote sensing imagery\n\n- [How Airbus Detects Anomalies in ISS Telemetry Data Using TFX](https://blog.tensorflow.org/2020/04/how-airbus-detects-anomalies-iss-telemetry-data-tfx.html) -\u003e uses an autoencoder\n\n* [AgriSen-COG](https://github.com/tselea/agrisen-cog) -\u003e a Multicountry, Multitemporal Large-Scale Sentinel-2 Benchmark Dataset for Crop Mapping: includes an anomaly detection preprocessing step\n\n#\n## Image retrieval\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/retrieval.png\" width=\"550\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eIllustration of the remote sensing image retrieval process.\u003c/b\u003e\n\u003c/p\u003e\n\nImage retrieval is the task of retrieving images from a collection that are similar to a query image. Image retrieval plays a vital role in remote sensing by enabling the efficient and effective search for relevant images from large image archives, and by providing a way to quantify changes in the environment over time. [Image source](https://www.mdpi.com/2072-4292/12/2/219)\n\n- [Demo_AHCL_for_TGRS2022](https://github.com/weiweisong415/Demo_AHCL_for_TGRS2022) -\u003e Asymmetric Hash Code Learning (AHCL) for remote sensing image retrieval\n\n- [GaLR](https://github.com/xiaoyuan1996/GaLR) -\u003e Remote Sensing Cross-Modal Text-Image Retrieval Based on Global and Local Information\n\n- [retrievalSystem](https://github.com/xiaoyuan1996/retrievalSystem) -\u003e cross-modal image retrieval system\n\n- [AMFMN](https://github.com/xiaoyuan1996/AMFMN) -\u003e Exploring a Fine-grained Multiscale Method for Cross-modal Remote Sensing Image Retrieval\n\n- [Active-Learning-for-Remote-Sensing-Image-Retrieval](https://github.com/flateon/Active-Learning-for-Remote-Sensing-Image-Retrieval) -\u003e unofficial implementation of paper: A Novel Active Learning Method in Relevance Feedback for Content-Based Remote Sensing Image Retrieval\n\n- [CMIR-NET](https://github.com/ushasi/CMIR-NET-A-deep-learning-based-model-for-cross-modal-retrieval-in-remote-sensing) -\u003e A deep learning based model for cross-modal retrieval in remote sensing\n\n- [Deep-Hash-learning-for-Remote-Sensing-Image-Retrieval](https://github.com/smallsmallflypigtang/Deep-Hash-learning-for-Remote-Sensing-Image-Retrieval) -\u003e Deep Hash Learning for Remote Sensing Image Retrieval\n\n- [MHCLN](https://github.com/MLEnthusiast/MHCLN) -\u003e Deep Metric and Hash-Code Learning for Content-Based Retrieval of Remote Sensing Images\n\n- [HydroViet_VOR](https://github.com/lannguyen0910/HydroViet_VOR) -\u003e Object Retrieval in satellite images with Triplet Network\n\n- [AMFMN](https://github.com/AICyberTeam/AMFMN) -\u003e Exploring a Fine-Grained Multiscale Method for Cross-Modal Remote Sensing Image Retrieval\n\n- [remote-sensing-image-retrieval](https://github.com/IBM/remote-sensing-image-retrieval) -\u003e Multi-Spectral Remote Sensing Image Retrieval using Geospatial Foundation Models (IBM Prithvi)\n\n- [Composed Image Retrieval for Remote Sensing](https://github.com/billpsomas/rscir)\n\n- [CSMAE](https://github.com/jakhac/CSMAE) -\u003e About\nCross-Sensor Masked Autoencoder for Content Based Image Retrieval in Remote Sensing\n\n#\n## Image Captioning\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/captioned.png\" width=\"600\"\u003e\n  \u003cbr\u003e\n  \u003cb\u003eExample captioned image.\u003c/b\u003e\n\u003c/p\u003e\n\nImage Captioning is the task of automatically generating a textual description of an image. In remote sensing, image captioning can be used to automatically generate captions for satellite or aerial images, which can be useful for a variety of purposes, such as image search and retrieval, data cataloging, and data dissemination. The generated captions can provide valuable information about the content of the images, including the location, the type of terrain or objects present, and the weather conditions, among others. This information can be used to quickly and easily understand the content of the images, without having to manually examine each image. [Image source](https://github.com/chan64/remote_sensing_image_captioning)\n\n- [awesome-remote-image-captioning](https://github.com/iOPENCap/awesome-remote-image-captioning) -\u003e a list of awesome remote sensing image captioning resources\n\n- [awesome-vision-language-models-for-earth-observation](https://github.com/geoaigroup/awesome-vision-language-models-for-earth-observation)\n\n- [CapFormer](https://github.com/Junjue-Wang/CapFormer) -\u003e Pure transformer for remote sensing image caption\n\n- [remote_sensing_image_captioning](https://github.com/chan64/remote_sensing_image_captioning) -\u003e Region Driven Remote Sensing Image Captioning\n\n- [Remote Sensing Image Captioning with Transformer and Multilabel Classification](https://github.com/hiteshK03/Remote-sensing-image-captioning-with-transformer-and-multilabel-classification)\n\n- [Siamese-spatial-Graph-Convolution-Network](https://github.com/ushasi/Siamese-spatial-Graph-Convolution-Network) -\u003e Siamese graph convolutional network for content based remote sensing image retrieval\n\n- [MLAT](https://github.com/Chen-Yang-Liu/MLAT) -\u003e Remote-Sensing Image Captioning Based on Multilayer Aggregated Transformer\n\n- [WordSent](https://github.com/hw2hwei/WordSent) -\u003e Word–Sentence Framework for Remote Sensing Image Captioning\n\n- [a-mask-guided-transformer-with-topic-token](https://github.com/Meditation0119/a-mask-guided-transformer-with-topic-token-for-remote-sensing-image-captioning) -\u003e A Mask-Guided Transformer Network with Topic Token for Remote Sensing Image Captioning\n\n- [Meta captioning](https://github.com/QiaoqiaoYang/MetaCaptioning) -\u003e A meta learning based remote sensing image captioning framework\n\n- [Transformer-for-image-captioning](https://github.com/RicRicci22/Transformer-for-image-captioning) -\u003e a transformer for image captioning, trained on the UCM dataset\n\n- [remote-sensing-image-caption](https://github.com/TalentBoy2333/remote-sensing-image-caption) -\u003e image classification and image caption by PyTorch\n\n- [Fine tuning CLIP with Remote Sensing (Satellite) images and captions](https://huggingface.co/blog/fine-tune-clip-rsicd) -\u003e fine tuning CLIP on the [RSICD](https://github.com/201528014227051/RSICD_optimal) image captioning dataset, to enable querying large catalogues in natural language. With [repo](https://github.com/arampacha/CLIP-rsicd), uses 🤗. Also read [Why and How to Fine-tune CLIP](https://dienhoa.github.io/dhblog/posts/finetune_clip.html)\n\n#\n## Visual Question Answering\n\nVisual Question Answering (VQA) is the task of automatically answering a natural language question about an image. In remote sensing, VQA enables users to interact with the images and retrieve information using natural language questions. For example, a user could ask a VQA system questions such as \"What is the type of land cover in this area?\", \"What is the dominant crop in this region?\" or \"What is the size of the city in this image?\". The system would then analyze the image and generate an answer based on its understanding of the image content.\n\n- [VQA-easy2hard](https://gitlab.lrz.de/ai4eo/reasoning/VQA-easy2hard) -\u003e From Easy to Hard: Learning Language-guided Curriculum for Visual Question Answering on Remote Sensing Data\n\n- [lit4rsvqa](https://git.tu-berlin.de/rsim/lit4rsvqa) -\u003e LiT-4-RSVQA: Lightweight Transformer-based Visual Question Answering in Remote Sensing\n\n- [Change-Agent](https://github.com/Chen-Yang-Liu/Change-Agent) -\u003e Towards Interactive Comprehensive Remote Sensing Change Interpretation and Analysis\n\n#\n## Mixed data learning\nMixed data learning is the process of learning from datasets that may contain an mix of images, textual and numeric data. Mixed data learning can help improve the accuracy of models by allowing them to learn from multiple sources at once and use more sophisticated methods to identify patterns and correlations.\n\n- [Multi-Input Deep Neural Networks with PyTorch-Lightning - Combine Image and Tabular Data](https://rosenfelder.ai/multi-input-neural-network-pytorch/) -\u003e excellent intro article using pytorch, not actually applied to satellite data but to real estate data, with [repo](https://github.com/MarkusRosen/pytorch_multi_input_example)\n\n- [Joint Learning from Earth Observation and OpenStreetMap Data to Get Faster Better Semantic Maps](https://arxiv.org/abs/1705.06057) -\u003e fusion based architectures and coarse-to-fine segmentation to include the OpenStreetMap layer into multispectral-based deep fully convolutional networks, arxiv paper\n\n- [pyimagesearch article on mixed-data](https://www.pyimagesearch.com/2019/02/04/keras-multiple-inputs-and-mixed-data/)\n\n- [pytorch-widedeep](https://github.com/jrzaurin/pytorch-widedeep) -\u003e A flexible package for multimodal-deep-learning to combine tabular data with text and images using Wide and Deep models in Pytorch\n\n- [accidentRiskMap](https://github.com/songtaohe/accidentRiskMap) -\u003e Inferring high-resolution traffic accident risk maps based on satellite imagery and GPS trajectories\n\n- [Sub-meter resolution canopy height map by Meta](https://research.facebook.com/blog/2023/4/every-tree-counts-large-scale-mapping-of-canopy-height-at-the-resolution-of-individual-trees/) -\u003e Satellite Metadata combined with outputs from simple CNN to regress canopy height\n\n- [methane-emission-project](https://github.com/stlbnmaria/methane-emission-project) -\u003e Classification CNNs was combined in an ensemble approach with traditional methods on tabular data\n\n#\n## Few \u0026 zero shot learning\nThis is a class of techniques which attempt to make predictions for classes with few, one or even zero examples provided during training. In zero shot learning (ZSL) the model is assisted by the provision of auxiliary information which typically consists of descriptions/semantic attributes/word embeddings for both the seen and unseen classes at train time ([ref](https://learnopencv.com/zero-shot-learning-an-introduction/)). These approaches are particularly relevant to remote sensing, where there may be many examples of common classes, but few or even zero examples for other classes of interest.\n\n- [Aerial-SAM](https://github.com/geoaigroup/Aerial-SAM) -\u003e Zero-Shot Refinement of Buildings’ Segmentation Models using SAM\n\n- [FSODM](https://github.com/lixiang-ucas/FSODM) -\u003e Few-shot Object Detection on Remote Sensing Images\n\n- [Few-Shot Classification of Aerial Scene Images via Meta-Learning](https://www.mdpi.com/2072-4292/13/1/108/htm) -\u003e 2020 publication, a classification model that can quickly adapt to unseen categories using only a few labeled samples\n\n- [Papers about Few-shot Learning / Meta-Learning on Remote Sensing](https://github.com/lx709/Few-shot-Learning-Meta-Learning-on-Remote-Sensing-Papers)\n\n- [SPNet](https://github.com/zoraup/SPNet) -\u003e Siamese-Prototype Network for Few-Shot Remote Sensing Image Scene Classification\n\n- [MDL4OW](https://github.com/sjliu68/MDL4OW) -\u003e Few-Shot Hyperspectral Image Classification With Unknown Classes Using Multitask Deep Learning\n\n- [P-CNN](https://github.com/Ybowei/P-CNN) -\u003e Prototype-CNN for Few-Shot Object Detection in Remote Sensing Images\n\n- [CIR-FSD-2022](https://github.com/Li-ZK/CIR-FSD-2022) -\u003e Context Information Refinement for Few-Shot Object Detection in Remote Sensing Images\n\n- [IEEE_TNNLS_Gia-CFSL](https://github.com/YuxiangZhang-BIT/IEEE_TNNLS_Gia-CFSL) -\u003e Graph Information Aggregation Cross-Domain Few-Shot Learning for Hyperspectral Image Classification\n\n- [TIP_2022_CMFSL](https://github.com/B-Xi/TIP_2022_CMFSL) -\u003e Few-shot Learning with Class-Covariance Metric for Hyperspectral Image Classification\n\n- [sen12ms-human-few-shot-classifier](https://github.com/MarcCoru/sen12ms-human-few-shot-classifier) -\u003e Humans are poor few-shot classifiers for Sentinel-2 land cover\n\n- [S3Net](https://github.com/ZhaohuiXue/S3Net) -\u003e S3Net: Spectral–Spatial Siamese Network for Few-Shot Hyperspectral Image Classification\n\n- [SiameseNet-for-few-shot-Hyperspectral-Classification](https://github.com/jjwwczy/jjwwczy-SiameseNet-for-few-shot-Hyperspectral-Classification) -\u003e 3DCSN:SiameseNet-for-few-shot-Hyperspectral-Classification\n\n- [MESSL](https://github.com/OMEGAFSL/MESSL) -\u003e Multiform Ensemble Self-Supervised Learning for Few-Shot Remote Sensing Scene Classification\n\n- [SCCNet](https://github.com/linhanwang/SCCNet) -\u003e Self-Correlation and Cross-Correlation Learning for Few-Shot Remote Sensing Image Semantic Segmentation\n\n- [OEM-Fewshot-Challenge](https://github.com/cliffbb/OEM-Fewshot-Challenge) -\u003e OpenEarthMap Land Cover Mapping Few-Shot Challenge\nGeneralized Few-shot Semantic Segmentation\n\n- [meteor](https://github.com/MarcCoru/meteor) -\u003e a small deep learning meta-model with a single output\n\n- [SegLand](https://github.com/LiZhuoHong/SegLand) -\u003e Generalized Few-Shot Meets Remote Sensing: Discovering Novel Classes in Land Cover Mapping via Hybrid Semantic Segmentation Framework. 1st place in the OpenEarthMap Land Cover Mapping Few-Shot Challenge\n\n- [MCFA-Pytorch](https://github.com/masuqiang/MCFA-Pytorch) -\u003e Multi-level Cross-modal Feature Alignment via Contrastive Learning towards Zero-shot Classification of Remote Sensing Image Scenes [paper](https://arxiv.org/abs/2306.06066)\n\n#\n## Self-supervised, unsupervised \u0026 contrastive learning\nSelf-supervised, unsupervised \u0026 contrastive learning are all methods of machine learning that use unlabeled data to train algorithms. Self-supervised learning uses labeled data to create an artificial supervisor, while unsupervised learning uses only the data itself to identify patterns and similarities. Contrastive learning uses pairs of data points to learn representations of data, usually for classification tasks. Note that self-supervised approaches are commonly used in the training of so-called Foundational models, since they enable learning from large quantities of unlablleded data, tyipcally time series.\n\n- [Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data](https://devblog.pytorchlightning.ai/seasonal-contrast-transferable-visual-representations-for-remote-sensing-73a17863ed07) -\u003e Seasonal Contrast (SeCo) is an effective pipeline to leverage unlabeled data for in-domain pre-training of remote sensing representations. Models trained with SeCo achieve better performance than their ImageNet pre-trained counterparts and state-of-the-art self-supervised learning methods on multiple downstream tasks. [paper](https://arxiv.org/abs/2103.16607) and [repo](https://github.com/ElementAI/seasonal-contrast)\n\n- [Unsupervised Learning for Land Cover Classification in Satellite Imagery](https://omdena.com/blog/land-cover-classification/)\n\n- [Tile2Vec: Unsupervised representation learning for spatially distributed data](https://ermongroup.github.io/blog/tile2vec/)\n\n- [Contrastive Sensor Fusion](https://github.com/descarteslabs/contrastive_sensor_fusion) -\u003e Code implementing Contrastive Sensor Fusion, an approach for unsupervised learning of multi-sensor representations targeted at remote sensing imagery\n\n- [hyperspectral-autoencoders](https://github.com/lloydwindrim/hyperspectral-autoencoders) -\u003e Tools for training and using unsupervised autoencoders and supervised deep learning classifiers for hyperspectral data, built on tensorflow. Autoencoders are unsupervised neural networks that are useful for a range of applications such as unsupervised feature learning and dimensionality reduction.\n\n- [MARTA GANs: Unsupervised Representation Learning for Remote Sensing Image Classification](https://github.com/BUPTLdy/MARTA-GAN)\n\n- [A generalizable and accessible approach to machine learning with global satellite imagery](https://www.nature.com/articles/s41467-021-24638-z) nature publication -\u003e MOSAIKS is designed to solve an unlimited number of tasks at planet-scale quickly using feature vectors, [with repo](https://github.com/Global-Policy-Lab/mosaiks-paper). Also see [mosaiks-api](https://github.com/calebrob6/mosaiks-api)\n\n- [contrastive-satellite](https://github.com/hakeemtfrank/contrastive-satellite) -\u003e Using contrastive learning to create embeddings from optical EuroSAT Satellite-2 imagery\n\n- [Self-Supervised Learning of Remote Sensing Scene Representations Using Contrastive Multiview Coding](https://github.com/vladan-stojnic/CMC-RSSR)\n\n- [Self-Supervised-Learner by spaceml-org](https://github.com/spaceml-org/Self-Supervised-Learner) -\u003e train a classifier with fewer labeled examples needed using self-supervised learning, example applied to UC Merced land use dataset\n\n- [deepsentinel](https://github.com/Lkruitwagen/deepsentinel) -\u003e a sentinel-1 and -2 self-supervised sensor fusion model for general purpose semantic embedding\n\n- [contrastive_SSL_ship_detection](https://github.com/alina2204/contrastive_SSL_ship_detection) -\u003e Contrastive self supervised learning for ship detection in Sentinel 2 images\n\n- [geography-aware-ssl](https://github.com/sustainlab-group/geography-aware-ssl) -\u003e uses spatially aligned images over time to construct temporal positive pairs in contrastive learning and geo-location to design pre-text tasks\n\n- [CNN-Supervised Classification](https://github.com/geojames/CNN-Supervised-Classification) -\u003e Python code for self-supervised classification of remotely sensed imagery - part of the Deep Riverscapes project\n\n- [clustimage](https://github.com/erdogant/clustimage) -\u003e a python package for unsupervised clustering of images\n\n- [LandSurfaceClustering](https://github.com/lhalloran/LandSurfaceClustering) -\u003e Land surface classification using remote sensing data with unsupervised machine learning (k-means)\n\n- [K-Means Clustering for Surface Segmentation of Satellite Images](https://medium.com/@maxfieldeland/k-means-clustering-for-surface-segmentation-of-satellite-images-ad1902791ebf)\n\n- [Sentinel-2 satellite imagery for crop classification using unsupervised clustering](https://medium.com/devseed/sentinel-2-satellite-imagery-for-crop-classification-part-2-47db3745eb49) -\u003e label groups of pixels based on temporal trends of their NDVI values\n\n- [TheColorOutOfSpace](https://github.com/stevinc/TheColorOutOfSpace) -\u003e The color out of space: learning self-supervised representations for Earth Observation imagery, using the BigEarthNet dataset\n\n- [Semantic segmentation of SAR images using a self supervised technique](https://github.com/cattale93/pytorch_self_supervised_learning)\n\n- [STEGO](https://github.com/mhamilton723/STEGO) -\u003e Unsupervised Semantic Segmentation by Distilling Feature Correspondences, with [paper](https://arxiv.org/abs/2203.08414)\n\n- [Unsupervised Segmentation of Hyperspectral Remote Sensing Images with Superpixels](https://github.com/mpBarbato/Unsupervised-Segmentation-of-Hyperspectral-Remote-Sensing-Images-with-Superpixels)\n\n- [SoundingEarth](https://github.com/khdlr/SoundingEarth) -\u003e Self-supervised Audiovisual Representation Learning for Remote Sensing Data, uses the SoundingEarth [Dataset](https://zenodo.org/record/5600379#.Yom4W5PMK3I)\n\n- [singleSceneSemSegTgrs2022](https://github.com/sudipansaha/singleSceneSemSegTgrs2022) -\u003e Unsupervised Single-Scene Semantic Segmentation for Earth Observation\n\n- [SSLRemoteSensing](https://github.com/flyakon/SSLRemoteSensing) -\u003e Semantic Segmentation of Remote Sensing Images With Self-Supervised Multitask Representation Learning\n\n- [CBT](https://github.com/VMarsocci/CBT) -\u003e Continual Barlow Twins: continual self-supervised learning for remote sensing semantic segmentation\n\n- [Unsupervised Satellite Image Classification based on Partial Adversarial Domain Adaptation](https://github.com/lwpyh/Unsupervised-Satellite-Image-Classfication-based-on-Partial-Domain-Adaptation) -\u003e Code for course project\n\n- [T2FTS](https://github.com/wdzhao123/T2FTS) -\u003e Teaching Teachers First and Then Student: Hierarchical Distillation to Improve Long-Tailed Object Recognition in Aerial Images\n\n- [SSLTransformerRS](https://github.com/HSG-AIML/SSLTransformerRS) -\u003e Self-supervised Vision Transformers for Land-cover Segmentation and\n  Classification\n\n- [DINO-MM](https://github.com/zhu-xlab/DINO-MM) -\u003e Self-supervised Vision Transformers for Joint SAR-optical Representation Learning\n\n- [SSL4EO-S12](https://github.com/zhu-xlab/SSL4EO-S12) -\u003e a large-scale dataset for self-supervised learning in Earth observation\n\n- [SSL4EO-Review](https://github.com/zhu-xlab/SSL4EO-Review) -\u003e Self-supervised Learning in Remote Sensing: A Review\n\n- [transfer_learning_cspt](https://github.com/ZhAnGToNG1/transfer_learning_cspt) -\u003e Consecutive Pretraining: A Knowledge Transfer Learning Strategy with Relevant Unlabeled Data for Remote Sensing Domain\n\n- [OTL](https://github.com/qlilx/OTL) -\u003e Clustering-Based Representation Learning through Output Translation and Its Application to Remote-Sensing Images\n\n- [Push-and-Pull-Network](https://github.com/WindVChen/Push-and-Pull-Network) -\u003e Contrastive Learning for Fine-grained Ship Classification in Remote Sensing Images\n\n- [vissl_experiments](https://github.com/lewfish/ssl/tree/main/vissl_experiments) -\u003e Self-supervised Learning using Facebook [VISSL](https://github.com/facebookresearch/vissl) on the RESISC-45 satellite imagery classification dataset\n\n- [MS2A-Net](https://github.com/Kasra2020/MS2A-Net) -\u003e MS 2 A-Net: Multi-scale spectral-spatial association network for hyperspectral image clustering\n\n- [UDA_for_RS](https://github.com/Levantespot/UDA_for_RS) -\u003e Unsupervised Domain Adaptation for Remote Sensing Semantic Segmentation with Transformer\n\n- [pytorch-ssl-building_extract](https://github.com/Chendeyue/pytorch-ssl-building_extract) -\u003e Research on Self-Supervised Building Information Extraction with High-Resolution Remote Sensing Images for Photovoltaic Potential Evaluation\n\n- [self-rare-wildlife](https://github.com/xcvil/self-rare-wildlife) -\u003e Self-Supervised Pretraining and Controlled Augmentation Improve Rare Wildlife Recognition in UAV Images\n\n- [SatMAE](https://github.com/sustainlab-group/SatMAE) -\u003e SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery\n\n- [FireCLR-Wildfires](https://github.com/spaceml-org/FireCLR-Wildfires) -\u003e Unsupervised Wildfire Change Detection based on Contrastive Learning\n\n- [FALSE](https://github.com/GeoX-Lab/FALSE) -\u003e False Negative Samples Aware Contrastive Learning for Semantic Segmentation of High-Resolution Remote Sensing Image\n\n- [MATTER](https://github.com/periakiva/MATTER) -\u003e Self-Supervised Material and Texture Representation Learning for Remote Sensing Tasks\n\n- [FGMAE](https://github.com/zhu-xlab/FGMAE) -\u003e Feature guided masked Autoencoder for self-supervised learning in remote sensing\n\n- [GFM](https://github.com/mmendiet/GFM) -\u003e Towards Geospatial Foundation Models via Continual Pretraining\n\n- [SatViT](https://github.com/antofuller/SatViT) -\u003e self-supervised training of multispectral optical and SAR vision transformers\n\n- [SITS-MoCo](https://github.com/YXu556/SITS-MoCo) -\u003e Self-supervised pre-training for large-scale crop mapping using Sentinel-2 time series\n\n- [DINO-MC](https://github.com/WennyXY/DINO-MC) -\u003e DINO-MC: Self-supervised Contrastive Learning for Remote Sensing Imagery with Multi-sized Local Crops\n\n- [official-CMID](https://github.com/NJU-LHRS/official-CMID) -\u003e A Unified Self-Supervised Learning Framework for Remote Sensing Image Understanding [paper](https://arxiv.org/abs/2304.09670)\n\n- [Domain-Adaptable-Self-Supervised-Representation-Learning-on-Remote-Sensing-Satellite-Imagery](https://github.com/muskaan712/Domain-Adaptable-Self-Supervised-Representation-Learning-on-Remote-Sensing-Satellite-Imagery) -\u003e Domain Adaptable Self-supervised Representation Learning on Remote Sensing Satellite Imagery\n\n- [PLRDiff](https://github.com/earth-insights/PLRDiff) -\u003e Unsupervised Hyperspectral Pansharpening via Low-rank Diffusion Model (Information Fusion 2024)\n\n- [RSOS_I2I](https://github.com/Sarmadfismael/RSOS_I2I) -\u003e Unsupervised Domain Adaptation for the Semantic Segmentation of Remote Sensing Images via One-Shot Image-to-Image Translation\n\n- [aws-smsl-geospatial-analysis-deforestation](https://github.com/aws-samples/aws-smsl-geospatial-analysis-deforestation) -\u003e Detecting deforestation using unsupervised K-means clustering on Sentinel-2 satellite imagery and SageMaker Studio Lab(SMSL) [Sagemaker notebook](https://studiolab.sagemaker.aws/import/github.com/aws-samples/aws-smsl-geospatial-analysis-deforestation/blob/main/geospatial_analysis_deforestation.ipynb)\n\n- [dinov2-remote-sensing](https://github.com/chagmgang/dinov2-remote-sensing) -\u003e Pytorch implementation and pretrained models for DINO v2 in remote sensing.\n\n#\n## Weakly \u0026 semi-supervised learning\n\nWeakly \u0026 semi-supervised learning are two methods of machine learning that use both labeled and unlabeled data for training. Weakly supervised learning uses weakly labeled data, which may be incomplete or inaccurate, while semi-supervised learning uses both labeled and unlabeled data. Weakly supervised learning is typically used in situations where labeled data is scarce and unlabeled data is abundant. Semi-supervised learning is typically used in situations where labeled data is abundant but also contains some noise or errors. Both techniques can be used to improve the accuracy of machine learning models by making use of additional data sources.\n\n- [MARE](https://github.com/VMarsocci/MARE) -\u003e self-supervised Multi-Attention REsu-net for semantic segmentation in remote sensing\n\n- [SSGF-for-HRRS-scene-classification](https://github.com/weihancug/SSGF-for-HRRS-scene-classification) -\u003e A semi-supervised generative framework with deep learning features for high-resolution remote sensing image scene classification\n\n- [SFGAN](https://github.com/MLEnthusiast/SFGAN) -\u003e Semantic-Fusion Gans for Semi-Supervised Satellite Image Classification\n\n- [SSDAN](https://github.com/alhichri/SSDAN) -\u003e Multi-Source Semi-Supervised Domain Adaptation Network for Remote Sensing Scene Classification\n\n- [HR-S2DML](https://github.com/jiankang1991/HR-S2DML) -\u003e High-Rankness Regularized Semi-Supervised Deep Metric Learning for Remote Sensing Imagery\n\n- [Semantic Segmentation of Satellite Images Using Point Supervision](https://github.com/KambachJannis/MasterThesis)\n\n- [fcd](https://github.com/jnyborg/fcd) -\u003e Fixed-Point GAN for Cloud Detection. A weakly-supervised approach, training with only image-level labels\n\n- [weak-segmentation](https://github.com/LendelTheGreat/weak-segmentation) -\u003e Weakly supervised semantic segmentation for aerial images in pytorch\n\n- [TNNLS_2022_X-GPN](https://github.com/B-Xi/TNNLS_2022_X-GPN) -\u003e Semisupervised Cross-scale Graph Prototypical Network for Hyperspectral Image Classification\n\n- [weakly_supervised](https://github.com/LobellLab/weakly_supervised) -\u003e Weakly Supervised Deep Learning for Segmentation of Remote Sensing Imagery. Demonstrates that segmentation can be performed using small datasets comprised of pixel or image labels\n\n- [wan](https://github.com/engrjavediqbal/wan) -\u003e Weakly-Supervised Domain Adaptation for Built-up Region Segmentation in Aerial and Satellite Imagery\n\n- [sourcerer](https://github.com/benjaminmlucas/sourcerer) -\u003e A Bayesian-inspired deep learning method for semi-supervised domain adaptation designed for land cover mapping from satellite image time series (SITS)\n\n- [MSMatch](https://github.com/gomezzz/MSMatch) -\u003e Semi-Supervised Multispectral Scene Classification with Few Labels. Includes code to work with both the RGB and the multispectral (MS) versions of EuroSAT dataset and the UC Merced Land Use (UCM) dataset\n\n- [Flood Segmentation on Sentinel-1 SAR Imagery with Semi-Supervised Learning](https://github.com/sidgan/ETCI-2021-Competition-on-Flood-Detection)\n\n- [Semi-supervised learning in satellite image classification](https://medium.com/sentinel-hub/semi-supervised-learning-in-satellite-image-classification-e0874a76fc61) -\u003e experimenting with MixMatch and the EuroSAT data set\n\n- [ScRoadExtractor](https://github.com/weiyao1996/ScRoadExtractor) -\u003e Scribble-based Weakly Supervised Deep Learning for Road Surface Extraction from Remote Sensing Images\n\n- [ICSS](https://github.com/alteia-ai/ICSS) -\u003e Weakly-supervised continual learning for class-incremental segmentation\n\n- [es-CP](https://github.com/majidseydgar/Res-CP) -\u003e Semi-Supervised Hyperspectral Image Classification Using a Probabilistic Pseudo-Label Generation Framework\n\n- [Flood_Mapping_SSL](https://github.com/YJ-He/Flood_Mapping_SSL) -\u003e Enhancement of Urban Floodwater Mapping From Aerial Imagery With Dense Shadows via Semisupervised Learning\n\n- [MS4D-Net-Building-Damage-Assessment](https://github.com/YJ-He/MS4D-Net-Building-Damage-Assessment) -\u003e MS4D-Net: Multitask-Based Semi-Supervised Semantic Segmentation Framework with Perturbed Dual Mean Teachers for Building Damage Assessment from High-Resolution Remote Sensing Imagery\n\n#\n## Active learning\n\nSupervised deep learning techniques typically require a huge number of annotated/labelled examples to provide a training dataset. However labelling at scale take significant time, expertise and resources. Active learning techniques aim to reduce the total amount of annotation that needs to be performed by selecting the most useful images to label from a large pool of unlabelled images, thus reducing the time to generate useful training datasets. These processes may be referred to as [Human-in-the-Loop Machine Learning](https://medium.com/pytorch/https-medium-com-robert-munro-active-learning-with-pytorch-2f3ee8ebec)\n\n- [Active learning for object detection in high-resolution satellite images](https://arxiv.org/abs/2101.02480)\n\n- [AIDE V2 - Tools for detecting wildlife in aerial images using active learning](https://github.com/microsoft/aerial_wildlife_detection)\n\n- [AstronomicAL](https://github.com/grant-m-s/AstronomicAL) -\u003e An interactive dashboard for visualisation, integration and classification of data using Active Learning\n\n- Follow tutorials for [active learning for object detection](https://docs.lightly.ai/docs/active-learning-yolov7) [and segmentation](https://docs.lightly.ai/docs/active-learning-for-driveable-area-segmentation-using-cityscapes) on the lightly platform.\n\n- [Active-Labeler by spaceml-org](https://github.com/spaceml-org/Active-Labeler) -\u003e a CLI Tool that facilitates labeling datasets with just a SINGLE line of code\n\n- [Labelling platform for Mapping Africa active learning project](https://github.com/agroimpacts/labeller)\n\n- [ChangeDetectionProject](https://github.com/previtus/ChangeDetectionProject) -\u003e Trying out Active Learning in with deep CNNs for Change detection on remote sensing data\n\n- [ALS4GAN](https://github.com/immuno121/ALS4GAN) -\u003e Active Learning for Improved Semi Supervised Semantic Segmentation in Satellite Images\n\n- [Active-Learning-for-Remote-Sensing-Image-Retrieval](https://github.com/flateon/Active-Learning-for-Remote-Sensing-Image-Retrieval) -\u003e unofficial implementation of paper: A Novel Active Learning Method in Relevance Feedback for Content-Based Remote Sensing Image Retrieval\n\n- [DIAL](https://github.com/alteia-ai/DIAL) -\u003e DIAL: Deep Interactive and Active Learning for Semantic Segmentation in Remote Sensing\n\n- [whales](https://github.com/microsoft/whales) -\u003e An active learning pipeline for identifying whales in high-resolution satellite imagery, by Microsoft\n\n- [AL4EO](https://github.com/Romain3Ch216/AL4EO) -\u003e a QGIS plug-in to run Active Learning techniques on Earth observation data\n\n#\n## Federated learning\n\nFederated learning is an approach to distributed machine learning where a central processor coordinates the training of an individual model in each of its clients. It is a type of distributed ML which means that the data is distributed among different devices or locations and the model is trained on all of them. The central processor aggregates the model updates from all the clients and then sends the global model parameters back to the clients. This is done to protect the privacy of data, as the data remains on the local device and only the global model parameters are shared with the central processor. This technique can be used to train models with large datasets that cannot be stored in a single device, as well as to enable certain privacy-preserving applications.\n\n- [Federated-Learning-for-Remote-Sensing](https://github.com/anandcu3/Federated-Learning-for-Remote-Sensing) -\u003e  implementation of three Federated Learning models\n\n- [Semantic-Segmentation-UNet-Federated](https://github.com/PratikGarai/Semantic-Segmentation-UNet-Federated) -\u003e FedUKD: Federated UNet Model with Knowledge Distillation for Land Use Classification from Satellite and Street Views\n\n- [MM-FL](https://git.tu-berlin.de/rsim/MM-FL) -\u003e Learning Across Decentralized Multi-Modal Remote Sensing Archives with Federated Learning\n\n#\n## Adversarial ML\n\nEfforts to detect falsified images \u0026 deepfakes\n\n- [UAE-RS](https://github.com/YonghaoXu/UAE-RS) -\u003e dataset that provides black-box adversarial samples in the remote sensing field\n\n- [PSGAN](https://github.com/xuxiangsun/PSGAN) -\u003e Perturbation Seeking Generative Adversarial Networks: A Defense Framework for Remote Sensing Image Scene Classification\n\n- [SACNet](https://github.com/YonghaoXu/SACNet) -\u003e Self-Attention Context Network: Addressing the Threat of Adversarial Attacks for Hyperspectral Image Classification\n\n#\n## Image registration\n\nImage registration is the process of registering one or more images onto another (typically well georeferenced) image. Traditionally this is performed manually by identifying control points (tie-points) in the images, for example using QGIS. This section lists approaches which mostly aim to automate this manual process. There is some overlap with the data fusion section but the distinction I make is that image registration is performed as a prerequisite to downstream processes which will use the registered data as an input.\n\n- [Wikipedia article on registration](https://en.wikipedia.org/wiki/Image_registration) -\u003e register for change detection or [image stitching](https://mono.software/2018/03/14/Image-stitching/)\n\n- [Phase correlation](https://en.wikipedia.org/wiki/Phase_correlation) is used to estimate the XY translation between two images with sub-pixel accuracy. Can be used for accurate registration of low resolution imagery onto high resolution imagery, or to register a [sub-image on a full image](https://www.mathworks.com/help/images/registering-an-image-using-normalized-cross-correlation.html) -\u003e Unlike many spatial-domain algorithms, the phase correlation method is resilient to noise, occlusions, and other defects. With [additional pre-processing](https://scikit-image.org/docs/dev/auto_examples/registration/plot_register_rotation.html) image rotation and scale changes can also be calculated.\n\n- [How to Co-Register Temporal Stacks of Satellite Images](https://medium.com/sentinel-hub/how-to-co-register-temporal-stacks-of-satellite-images-5167713b3e0b)\n\n- [image-matching-models](https://github.com/gmberton/image-matching-models) -\u003e  easily try 23 different image matching methods\n\n- [ImageRegistration](https://github.com/jandremarais/ImageRegistration) -\u003e Interview assignment for multimodal image registration using SIFT\n\n- [imreg_dft](https://github.com/matejak/imreg_dft) -\u003e Image registration using discrete Fourier transform. Given two images it can calculate the difference between scale, rotation and position of imaged features.\n\n- [arosics](https://danschef.git-pages.gfz-potsdam.de/arosics/doc/about.html) -\u003e Perform automatic subpixel co-registration of two satellite image datasets using phase-correlation, XY translations only.\n\n- [SubpixelAlignment](https://github.com/vldkhramtsov/SubpixelAlignment) -\u003e Implementation of tiff image alignment through phase correlation for pixel- and subpixel-bias\n\n- [cnn-registration](https://github.com/yzhq97/cnn-registration) -\u003e A image registration method using convolutional neural network features written in Python2, Tensorflow 1.5\n\n- [Siamese_ShiftNet](https://github.com/simon-donike/Siamese_ShiftNet) -\u003e NN predicting spatial coregistration shift of remote sensing imagery. Adapted from HighRes-net\n\n- [ImageCoregistration](https://github.com/ily-R/ImageCoregistration) -\u003e Image registration with openCV using sift and RANSAC\n\n- [mapalignment](https://github.com/Lydorn/mapalignment) -\u003e Aligning and Updating Cadaster Maps with Remote Sensing Images\n\n- [CVPR21-Deep-Lucas-Kanade-Homography](https://github.com/placeforyiming/CVPR21-Deep-Lucas-Kanade-Homography) -\u003e deep learning pipeline to accurately align challenging multimodality images. The method is based on traditional Lucas-Kanade algorithm with feature maps extracted by deep neural networks.\n\n- [eolearn](https://eo-learn.readthedocs.io/en/latest/_modules/eolearn/coregistration/coregistration.html) implements phase correlation, feature matching and [ECC](https://learnopencv.com/image-alignment-ecc-in-opencv-c-python/)\n\n- [Reprojecting the Perseverance landing footage onto satellite imagery](https://matthewearl.github.io/2021/03/06/mars2020-reproject/)\n\n- Kornia provides [image registration](https://kornia.readthedocs.io/en/latest/applications/image_registration.html)\n\n- [LoFTR](https://github.com/zju3dv/LoFTR) -\u003e Detector-Free Local Feature Matching with Transformers. Good performance matching satellite image pairs, tryout the web demo on your data\n\n- [image-to-db-registration](https://gitlab.orfeo-toolbox.org/remote_modules/image-to-db-registration) -\u003e This remote module implements an algorithm for automated vector Database registration onto an Image. Implemented in the orfeo-toolbox\n\n- [MS_HLMO_registration](https://github.com/MrPingQi/MS_HLMO_registration) -\u003e Multi-scale Histogram of Local Main Orientation for Remote Sensing Image Registration, with [paper](https://arxiv.org/abs/2204.00260)\n\n- [cnn-matching](https://github.com/lan-cz/cnn-matching) -\u003e Deep learning algorithm for feature matching of cross modality remote sensing images\n\n- [Imatch-P](https://github.com/geoyee/Imatch-P) -\u003e A demo using SuperGlue and SuperPoint to do the image matching task based PaddlePaddle\n\n- [NBR-Net](https://github.com/xuyingxiao/NBR-Net) -\u003e A Non-rigid Bi-directional Registration Network for Multi-temporal Remote Sensing Images\n\n- [MU-Net](https://github.com/woshiybc/Multi-Scale-Unsupervised-Framework-MSUF) -\u003e A Multi-Scale Framework with Unsupervised Learning for Remote Sensing Image Registration\n\n- [unsupervisedDeepHomographyRAL2018](https://github.com/tynguyen/unsupervisedDeepHomographyRAL2018) -\u003e Unsupervised Deep Homography applied to aerial data\n\n- [registration_cnn_ntg](https://github.com/zhangliukun/registration_cnn_ntg) -\u003e A Multispectral Image Registration Method Based on Unsupervised Learning\n\n- [remote-sensing-images-registration-dataset](https://github.com/liliangzhi110/remote-sensing-images-registration-dataset) -\u003e at 0.23m, 3.75m \u0026 30m resolution\n\n- [semantic-template-matching](https://github.com/liliangzhi110/semantictemplatematching) -\u003e A deep learning semantic template matching framework for remote sensing image registration\n\n- [GMN-Generative-Matching-Network](https://github.com/ei1994/GMN-Generative-Matching-Network) -\u003e Deep Generative Matching Network for Optical and SAR Image Registration\n\n- [SOMatch](https://github.com/system123/SOMatch) -\u003e A deep learning framework for matching of SAR and optical imagery\n\n- [Interspectral image registration dataset](https://medium.com/dronehub/datasets-96fc4f9a92e5) -\u003e including satellite and drone imagery\n\n- [RISG-image-matching](https://github.com/lan-cz/RISG-image-matching) -\u003e A rotation invariant SuperGlue image matching algorithm\n\n- [DeepAerialMatching_pytorch](https://github.com/jaehyunnn/DeepAerialMatching_pytorch) -\u003e A Two-Stream Symmetric Network with Bidirectional Ensemble for Aerial Image Matching\n\n- [DPCN](https://github.com/ZJU-Robotics-Lab/DPCN) -\u003e Deep Phase Correlation for End-to-End Heterogeneous Sensor Measurements Matching\n\n- [FSRA](https://github.com/Dmmm1997/FSRA) -\u003e A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localization\n\n- [IHN](https://github.com/imdumpl78/IHN) -\u003e Iterative Deep Homography Estimation\n\n- [OSMNet](https://github.com/zhanghan9718/OSMNet) -\u003e Explore Better Network Framework for High-Resolution Optical and SAR Image Matching\n\n- [L2_Siamese](https://github.com/TheKiteFlier/L2_Siamese) -\u003e Registration of Multiresolution Remote Sensing Images Based on L2-Siamese Model\n\n- [Multi-Step-Deformable-Registration](https://github.com/mpapadomanolaki/Multi-Step-Deformable-Registration) -\u003e Unsupervised Multi-Step Deformable Registration of Remote Sensing Imagery based on Deep Learning\n\n#\n## Terrain mapping, Disparity Estimation, Lidar, DEMs \u0026 NeRF\n\nMeasure surface contours \u0026 locate 3D points in space from 2D images. NeRF stands for Neural Radiance Fields and is the term used in deep learning communities to describe a model that generates views of complex 3D scenes based on a partial set of 2D images\n\n- [Wikipedia DEM article](https://en.wikipedia.org/wiki/Digital_elevation_model) and [phase correlation](https://en.wikipedia.org/wiki/Phase_correlation) article\n\n- [Intro to depth from stereo](https://github.com/IntelRealSense/librealsense/blob/master/doc/depth-from-stereo.md)\n\n- Map terrain from stereo images to produce a digital elevation model (DEM) -\u003e high resolution \u0026 paired images required, typically 0.3 m, e.g. [Worldview](https://dg-cms-uploads-production.s3.amazonaws.com/uploads/document/file/37/DG-WV2ELEVACCRCY-WP.pdf)\n\n- Process of creating a DEM [here](https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLI-B1/327/2016/isprs-archives-XLI-B1-327-2016.pdf)\n\n- [ArcGIS can generate DEMs from stereo images](http://pro.arcgis.com/en/pro-app/help/data/imagery/generate-elevation-data-using-the-dems-wizard.htm)\n\n- [S2P](https://github.com/centreborelli/s2p) -\u003e S2P is a Python library and command line tool that implements a stereo pipeline which produces elevation models from images taken by high resolution optical satellites such as Pléiades, WorldView, QuickBird, Spot or Ikonos.\n\n- [Predict the fate of glaciers](https://github.com/geohackweek/glacierhack_2018)\n\n- [monodepth - Unsupervised single image depth prediction with CNNs](https://github.com/mrharicot/monodepth)\n\n- [Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches](https://github.com/jzbontar/mc-cnn)\n\n- [Terrain and hydrological analysis based on LiDAR-derived digital elevation models (DEM) - Python package](https://github.com/giswqs/lidar)\n\n- [Phase correlation in scikit-image](https://scikit-image.org/docs/0.13.x/auto_examples/transform/plot_register_translation.html)\n\n- [3DCD](https://github.com/VMarsocci/3DCD) -\u003e Inferring 3D change detection from bitemporal optical images\n\n- [Reconstructing 3D buildings from aerial LiDAR with Mask R-CNN](https://medium.com/geoai/reconstructing-3d-buildings-from-aerial-lidar-with-ai-details-6a81cb3079c0)\n\n- [ResDepth](https://github.com/stuckerc/ResDepth) -\u003e A Deep Prior For 3D Reconstruction From High-resolution Satellite Images\n\n- [overhead-geopose-challenge](https://www.drivendata.org/competitions/78/overhead-geopose-challenge/) -\u003e competition to build computer vision algorithms that can effectively model the height and pose of ground objects for monocular satellite images taken from oblique angles. Blog post [MEET THE WINNERS OF THE OVERHEAD GEOPOSE CHALLENGE](https://www.drivendata.co/blog/overhead-geopose-challenge-winners/)\n\n- [cars](https://github.com/CNES/cars) -\u003e a dedicated and open source 3D tool to produce Digital Surface Models from satellite imaging by photogrammetry. This Multiview stereo pipeline is intended for massive DSM production with a robust and performant design\n\n- [ImageToDEM](https://github.com/Panagiotou/ImageToDEM) -\u003e Generating Elevation Surface from a Single RGB Remotely Sensed Image Using a U-Net for generator and a PatchGAN for the discriminator\n\n- [IMELE](https://github.com/speed8928/IMELE) -\u003e Building Height Estimation from Single-View Aerial Imagery\n\n- [ridges](https://github.com/mikeskaug/ridges) -\u003e deep semantic segmentation model for identifying ridges in topography\n\n- [planet_tools](https://github.com/disbr007/planet_tools) -\u003e Selection of imagery from Planet API for creation of stereo elevation models\n\n- [SatelliteNeRF](https://github.com/Kai-46/SatelliteNeRF) -\u003e PyTorch-based Neural Radiance Fields adapted to satellite domain\n\n- [SatelliteSfM](https://github.com/Kai-46/SatelliteSfM) -\u003e A library for solving the satellite structure from motion problem\n\n- [SatelliteSurfaceReconstruction](https://github.com/SBCV/SatelliteSurfaceReconstruction) -\u003e 3D Surface Reconstruction From Multi-Date Satellite Images, ISPRS, 2021\n\n- [son2sat](https://github.com/giovgiac/son2sat) -\u003e A neural network coded in TensorFlow 1 that produces satellite images from acoustic images\n\n- [aerial_mtl](https://github.com/marcelampc/aerial_mtl) -\u003e PyTorch implementation for multi-task learning with aerial images to learn both semantics and height from aerial image datasets; fuses RGB \u0026 lidar\n\n- [ReKlaSat-3D](https://github.com/MacOS/ReKlaSat-3D) -\u003e 3D Reconstruction and Classification from Very High Resolution Satellite Imagery\n\n- [M3Net](https://github.com/lauraset/BuildingHeightModel) -\u003e A deep learning method for building height estimation using high-resolution multi-view imagery over urban areas\n\n- [HMSM-Net](https://github.com/Sheng029/HMSM-Net) -\u003e Hierarchical multi-scale matching network for disparity estimation of high-resolution satellite stereo images\n\n- [StereoMatchingRemoteSensing](https://github.com/Sheng029/StereoMatchingRemoteSensing) -\u003e Dual-Scale Matching Network for Disparity Estimation of High-Resolution Remote Sensing Images\n\n- [satnerf](https://centreborelli.github.io/satnerf/) -\u003e Learning Multi-View Satellite Photogrammetry With Transient Objects and Shadow Modeling Using RPC Cameras\n\n- [SatMVS](https://github.com/WHU-GPCV/SatMVS) -\u003e Rational Polynomial Camera Model Warping for Deep Learning Based Satellite Multi-View Stereo Matching\n\n- [ImpliCity](https://github.com/prs-eth/ImpliCity) -\u003e reconstructs digital surface models (DSMs) from raw photogrammetric 3D point clouds and ortho-images with the help of an implicit neural 3D scene representation\n\n- [WHU-Stereo](https://github.com/Sheng029/WHU-Stereo) -\u003e a large-scale dataset for stereo matching of high-resolution satellite imagery \u0026 several deep learning methods for stereo matching. Methods include StereoNet, Pyramid Stereo Matching Network \u0026 HMSM-Net\n\n- [Photogrammetry-Guide](https://github.com/mikeroyal/Photogrammetry-Guide) -\u003e A guide covering Photogrammetry including the applications, libraries and tools that will make you a better and more efficient Photogrammetry development\n\n- [DSM-to-DTM](https://github.com/mdmeadows/DSM-to-DTM) -\u003e Exploring the use of machine learning to convert a Digital Surface Model (e.g. SRTM) to a Digital Terrain Model\n\n- [GF-7_Stereo_Matching](https://github.com/Sheng029/GF-7_Stereo_Matching) -\u003e Large Scene DSM Generation of Gaofen-7 Imagery Combined with Deep Learning\n\n- [Mapping drainage ditches in forested landscapes using deep learning and aerial laser scanning](https://github.com/williamlidberg/Mapping-drainage-ditches-in-forested-landscapes-using-deep-learning-and-aerial-laser-scanning)\n\n#\n## Thermal Infrared\n\nThermal infrared remote sensing is a technique used to detect and measure thermal radiation emitted from the Earth’s surface. This technique can be used to measure the temperature of the ground and any objects on it and can detect the presence of different materials. Thermal infrared remote sensing is used to assess land cover, detect land-use changes, and monitor urban heat islands, as well as to measure the temperature of the ground during nighttime or in areas of limited visibility.\n\n- [Object_Classification_in_Thermal_Images](https://www.researchgate.net/publication/328400392_Object_Classification_in_Thermal_Images_using_Convolutional_Neural_Networks_for_Search_and_Rescue_Missions_with_Unmanned_Aerial_Systems) -\u003e classification accuracy was improved by adding the object size as a feature directly within the CNN\n\n- [Thermal imaging with satellites](https://chrieke.medium.com/thermal-imaging-with-satellites-34f381856dd1) blog post by Christoph Rieke\n\n#\n## SAR\n\nSAR (synthetic aperture radar) is used to detect and measure the properties of objects and surfaces on the Earth's surface. SAR can be used to detect changes in terrain, features, and objects over time, as well as to measure the size, shape, and composition of objects and surfaces. SAR can also be used to measure moisture levels in soil and vegetation, or to detect and monitor changes in land use.\n\n- [awesome-sar](https://github.com/RadarCODE/awesome-sar)\n\n- [awesome-sar-deep-learning](https://github.com/neeraj3029/awesome-sar-deep-learning)\n\n- [MERLIN](https://gitlab.telecom-paris.fr/ring/MERLIN) -\u003e self-supervised training of deep despeckling networks with MERLIN\n\n- [PySAR - InSAR (Interferometric Synthetic Aperture Radar) timeseries analysis in python](https://github.com/hfattahi/PySAR)\n\n- [Synthetic Aperture Radar (SAR) Analysis With Clarifai](https://www.clarifai.com/blog/synthetic-aperture-radar-sar-analysis-with-clarifai)\n\n- [Labeled SAR imagery dataset of ten geophysical phenomena from Sentinel-1 wave mode](https://www.seanoe.org/data/00456/56796/) consists of more than 37,000 SAR vignettes divided into ten defined geophysical categories\n\n- [Implementing an Ensemble Convolutional Neural Network on Sentinel-1 Synthetic Aperture Radar data and Sentinel-3 Radiometric data for the detecting of forest fires](https://github.com/aalling93/ECNN-on-SAR-data-and-Radiometry-data)\n\n- [s1_parking_occupancy](https://github.com/sdrdis/s1_parking_occupancy) -\u003e PARKING OCCUPANCY ESTIMATION ON SENTINEL-1 IMAGES\n\n- [Experiments on Flood Segmentation on Sentinel-1 SAR Imagery with Cyclical Pseudo Labeling and Noisy Student Training](https://github.com/sidgan/ETCI-2021-Competition-on-Flood-Detection)\n\n- [SpaceNet_SAR_Buildings_Solutions](https://github.com/SpaceNetChallenge/SpaceNet_SAR_Buildings_Solutions) -\u003e The winning solutions for the SpaceNet 6 Challenge\n\n- [Mapping and monitoring of infrastructure in desert regions with Sentinel-1](https://github.com/ESA-PhiLab/infrastructure)\n\n- [xView3](https://iuu.xview.us/) is a competition to detect dark vessels using computer vision and global SAR satellite imagery. [First place solution](https://github.com/DIUx-xView/xView3_first_place) and [second place solution](https://github.com/DIUx-xView/xView3_second_place). Additional places up to fifth place are available at the [xView GitHub Organization page](https://github.com/DIUx-xView/)\n\n- [Winners of the STAC Overflow: Map Floodwater from Radar Imagery competition](https://github.com/drivendataorg/stac-overflow)\n\n- [deSpeckNet-TF-GEE](https://github.com/adugnag/deSpeckNet-TF-GEE) -\u003e deSpeckNet: Generalizing Deep Learning Based SAR Image Despeckling\n\n- [cnn_sar_image_classification](https://github.com/diogosens/cnn_sar_image_classification) -\u003e CNN for classifying SAR images of the Amazon Rainforest\n\n- [s1_icetype_cnn](https://github.com/nansencenter/s1_icetype_cnn) -\u003e Retrieve sea ice type from Sentinel-1 SAR with CNN\n\n- [MP-ResNet](https://github.com/ggsDing/SARSeg) -\u003e Multi-path Residual Network for the Semantic segmentation of PolSAR Images'\n\n- [TGRS_DisOptNet](https://github.com/jiankang1991/TGRS_DisOptNet) -\u003e Distilling Semantic Knowledge from Optical Images for Weather-independent Building Segmentation\n\n- [SAR_CD_DDNet](https://github.com/summitgao/SAR_CD_DDNet) -\u003e PyTorch implementation of Change Detection in Synthetic Aperture Radar Images Using a Dual Domain Network\n\n- [SAR_CD_MS_CapsNet](https://github.com/summitgao/SAR_CD_MS_CapsNet) -\u003e Change Detection in SAR Images Based on Multiscale Capsule Network\n\n- Toushka Waterbodies Segmentation from four different combinations of Sentinel-1 SAR imagery and Digital Elevation Model with Pytorch and U-net. -\u003e [code](https://github.com/MuhammedM294/waterseg)\n\n- [sar_transformer](https://github.com/malshaV/sar_transformer) -\u003e Transformer based SAR image despeckling, trained with synthetic imagery, with [paper](https://arxiv.org/abs/2201.09355)\n\n- [SSDD ship detection dataset](https://github.com/TianwenZhang0825/Official-SSDD)\n\n- [Semantic segmentation of SAR images using a self supervised technique](https://github.com/cattale93/pytorch_self_supervised_learning)\n\n- [Ship Detection on Remote Sensing Synthetic Aperture Radar Data](https://github.com/JasonManesis/Ship-Detection-on-Remote-Sensing-Synthetic-Aperture-Radar-Data) -\u003e based on the architectures of the Faster-RCNN and YOLOv5 networks\n\n- [Target Recognition in SAR](https://github.com/NateDiR/sar_target_recognition_deep_learning) -\u003e Identify Military Vehicles in Satellite Imagery with TensorFlow, with [article](https://python.plainenglish.io/identifying-military-vehicles-in-satellite-imagery-with-tensorflow-96015634129d)\n\n- [DSN](https://github.com/Alien9427/DSN) -\u003e Deep SAR-Net: Learning objects from signals\n\n- [SAR_denoising](https://github.com/MathieuRita/SAR_denoising) -\u003e project on application of FFDNet to SAR images\n\n- [cnninsar](https://github.com/subhayanmukherjee/cnninsar) -\u003e CNN-Based InSAR Denoising and Coherence Metric\n\n- [sar](https://github.com/GeomaticsAndRS/sar) -\u003e Despeckling Synthetic Aperture Radar Images using a Deep Residual CNN\n\n- [GCBANet](https://github.com/TianwenZhang0825/GCBANet) -\u003e A Global Context Boundary-Aware Network for SAR Ship Instance Segmentation\n\n- [SAR_CD_GKSNet](https://github.com/summitgao/SAR_CD_GKSNet) -\u003e Change Detection from Synthetic Aperture Radar Images via Graph-Based Knowledge Supplement Network\n\n- [pixel-wise-segmentation-of-sar](https://github.com/flyingshan/pixel-wise-segmentation-of-sar-imagery-using-encoder-decoder-network-and-fully-connected-crf) -\u003e Pixel-Wise Segmentation of SAR Imagery Using Encoder-Decoder Network and Fully-Connected CRF\n\n- [SAR_Ship_detection_CFAR](https://github.com/Rc-W024/SAR_Ship_detection_CFAR) -\u003e An improved two-parameter CFAR algorithm based on Rayleigh distribution and Mathematical Morphology for SAR ship detection\n\n- [sar_snow_melt_timing](https://github.com/egagli/sar_snow_melt_timing) -\u003e notebooks and tools to identify snowmelt timing using timeseries analysis of backscatter of Sentinel-1 C-band SAR\n\n- [Denoising radar satellite images using deep learning in Python](https://medium.com/@petebch/denoising-radar-satellite-images-using-deep-learning-in-python-946daad31022) -\u003e Medium article on [deepdespeckling](https://github.com/hi-paris/deepdespeckling)\n\n- [random-wetlands](https://github.com/ekcomputer/random-wetlands) -\u003e Random forest classification for wetland vegetation from synthetic aperture radar dataset\n\n- [AGSDNet](https://github.com/RTSIR/AGSDNet) -\u003e AGSDNet: Attention and Gradient-Based SAR Denoising Network\n\n- [LFG-Net](https://github.com/Evarray/LFG-Net) -\u003e LFG-Net: Low-Level Feature Guided Network for Precise Ship Instance Segmentation in SAR Images\n\n- [sar_sift](https://github.com/yishiliuhuasheng/sar_sift) -\u003e Image registration algorithm\n\n- [SAR-Despeckling](https://github.com/ImageRestorationToolbox/SAR-Despeckling) -\u003e toolbox\n\n- [cogsima2022](https://github.com/galatolofederico/cogsima2022) -\u003e Enhancing land subsidence awareness via InSAR data and Deep Transformers\n\n- [XAI4SAR-PGIL](https://github.com/Alien9427/XAI4SAR-PGIL) -\u003e Physically Explainable CNN for SAR Image Classification\n\n- [PolSARFormer](https://github.com/aj1365/PolSARFormer) -\u003e Local Window Attention Transformer for Polarimetric SAR Image Classification\n\n- [DC4Flood](https://github.com/Kasra2020/DC4Flood) -\u003e A deep clustering framework for rapid flood detection using Sentinel-1 SAR imagery\n\n- [Sentinel1-Flood-Finder](https://github.com/cordmaur/Sentinel1-Flood-Finder) -\u003e Flood Finder Package from Sentinel 1 Imagery\n\n- [bayes-forest-structure](https://github.com/prs-eth/bayes-forest-structure) -\u003e Country-wide Retrieval of Forest Structure From Optical and SAR Satellite Imagery With Bayesian Deep Learning [paper](https://www.sciencedirect.com/science/article/pii/S0924271622003045)\n\n#\n## NDVI - vegetation index\n\nNormalized Difference Vegetation Index (NDVI) is an index used to measure the amount of healthy vegetation in a given area. It is calculated by taking the difference between the near-infrared (NIR) and red (red) bands of a satellite image, and dividing by the sum of the two bands. NDVI can be used to identify areas of healthy vegetation and to assess the health of vegetation in a given area. `ndvi = np.true_divide((ir - r), (ir + r))`\n\n- [Example notebook local](http://nbviewer.jupyter.org/github/HyperionAnalytics/PyDataNYC2014/blob/master/ndvi_calculation.ipynb)\n\n- [Landsat data in cloud optimised (COG) format analysed for NDVI](https://github.com/pangeo-data/pangeo-example-notebooks/blob/master/landsat8-cog-ndvi.ipynb) with [medium article here](https://medium.com/pangeo/cloud-native-geoprocessing-of-earth-observation-satellite-data-with-pangeo-997692d91ca2).\n\n- [Identifying Buildings in Satellite Images with Machine Learning and Quilt](https://github.com/jyamaoka/LandUse) -\u003e NDVI \u0026 edge detection via gaussian blur as features, fed to TPOT for training with labels from OpenStreetMap, modelled as a two class problem, “Buildings” and “Nature”\n\n- [Seeing Through the Clouds - Predicting Vegetation Indices Using SAR](https://medium.com/descarteslabs-team/seeing-through-the-clouds-34a24f84b599)\n\n- [NDVI-Net](https://github.com/HaoZhang1018/NDVI-Net) -\u003e NDVI-Net: A fusion network for generating high-resolution normalized difference vegetation index in remote sensing\n\n- [Awesome-Vegetation-Index](https://github.com/px39n/Awesome-Vegetation-Index)\n\n- [Remote-Sensing-Indices-Derivation-Tool](https://github.com/rander38/Remote-Sensing-Indices-Derivation-Tool) -\u003e Calculate spectral remote sensing indices from satellite imagery\n\n#\n## General image quality\n\nImage quality describes the degree of accuracy with which an image can represent the original object. Image quality is typically measured by the amount of detail, sharpness, and contrast that an image contains. Factors that contribute to image quality include the resolution, format, and compression of the image.\n\n- [lvrnet](https://github.com/Achleshwar/lvrnet) -\u003e Lightweight Image Restoration for Aerial Images under Low Visibility\n\n- [jitter-compensation](https://github.com/caiya55/jitter-compensation) -\u003e Remote Sensing Image Jitter Detection and Compensation Using CNN\n\n- [DeblurGANv2](https://github.com/VITA-Group/DeblurGANv2) -\u003e Deblurring (Orders-of-Magnitude) Faster and Better\n\n- [image-quality-assessment](https://github.com/idealo/image-quality-assessment) -\u003e CNN to predict the aesthetic and technical quality of images\n\n- [DOTA-C](https://github.com/hehaodong530/DOTA-C) -\u003e evaluating the robustness of object detection models to 19 types of image quality degradation\n\n- [piq](https://github.com/photosynthesis-team/piq) -\u003e a collection of measures and metrics for image quality assessment\n\n- [FFA-Net](https://github.com/zhilin007/FFA-Net) -\u003e Feature Fusion Attention Network for Single Image Dehazing\n\n- [DeepCalib](https://github.com/alexvbogdan/DeepCalib) -\u003e A Deep Learning Approach for Automatic Intrinsic Calibration of Wide Field-of-View Cameras\n\n- [PerceptualSimilarity](https://github.com/richzhang/PerceptualSimilarity) -\u003e LPIPS is a perceptual metric which aims to overcome the limitations of traditional metrics such as PSNR \u0026 SSIM, to better represent the features the human eye picks up on\n\n- [Optical-RemoteSensing-Image-Resolution](https://github.com/wenjiaXu/Optical-RemoteSensing-Image-Resolution) -\u003e Deep Memory Connected Neural Network for Optical Remote Sensing Image Restoration. Two applications: Gaussian image denoising and single image super-resolution\n\n- [Hyperspectral-Deblurring-and-Destriping](https://github.com/ImageRestorationToolbox/Hyperspectral-Deblurring-and-Destriping)\n\n- [HyDe](https://github.com/Helmholtz-AI-Energy/HyDe) -\u003e Hyperspectral Denoising algorithm toolbox in Python\n\n- [HLF-DIP](https://github.com/Keiv4n/HLF-DIP) -\u003e Unsupervised Hyperspectral Denoising Based on Deep Image Prior and Least Favorable Distribution\n\n- [RQUNetVAE](https://github.com/trile83/RQUNetVAE) -\u003e Riesz-Quincunx-UNet Variational Auto-Encoder for Satellite Image Denoising\n\n- [deep-hs-prior](https://github.com/acecreamu/deep-hs-prior) -\u003e Deep Hyperspectral Prior: Denoising, Inpainting, Super-Resolution\n\n- [iquaflow](https://github.com/satellogic/iquaflow) -\u003e from Satellogic, an image quality framework that aims at providing a set of tools to assess image quality by using the performance of AI models trained on the images as a proxy.\n\n#\n## Synthetic data\n\nTraining data can be hard to acquire, particularly for rare events such as change detection after disasters, or imagery of rare classes of objects. In these situations, generating synthetic training data might be the only option. This has become quite sophisticated, with 3D models being use with open source games engines such as [Unreal](https://www.unrealengine.com/en-US/).\n\n- [The Synthinel-1 dataset: a collection of high resolution synthetic overhead imagery for building segmentation](https://arxiv.org/ftp/arxiv/papers/2001/2001.05130.pdf) with [repo](https://github.com/timqqt/Synthinel)\n\n- [RarePlanes](https://registry.opendata.aws/rareplanes/) -\u003e incorporates both real and synthetically generated satellite imagery including aircraft. Read the [arxiv paper](https://arxiv.org/abs/2006.02963) and checkout [this repo](https://github.com/jdc08161063/RarePlanes). Note the dataset is available through the AWS Open-Data Program for free download\n\n- Read [this article from NVIDIA](https://developer.nvidia.com/blog/preparing-models-for-object-detection-with-real-and-synthetic-data-and-tao-toolkit/) which discusses fine tuning a model pre-trained on synthetic data (Rareplanes) with 10% real data, then pruning the model to reduce its size, before quantizing the model to improve inference speed\n\n- [BlenderGIS](https://github.com/domlysz/BlenderGIS) could be used for synthetic data generation\n\n- [bifrost.ai](https://www.bifrost.ai/) -\u003e simulated data service with geospatial output data formats\n\n- [oktal-se](https://www.oktal-se.fr/deep-learning/) -\u003e software for generating simulated data across a wide range of bands including optical and SAR\n\n- [rendered.ai](https://rendered.ai/) -\u003e The Platform as a Service for Creating Synthetic Data\n\n- [synthetic_xview_airplanes](https://github.com/yangxu351/synthetic_xview_airplanes) -\u003e creation of airplanes synthetic dataset using ArcGIS CityEngine\n\n- [Import OpenStreetMap data into Unreal Engine 4](https://github.com/ue4plugins/StreetMap)\n\n- [deepfake-satellite-images](https://github.com/RijulGupta-DM/deepfake-satellite-images) -\u003e dataset that includes over 1M images of synthetic aerial images\n\n- [synthetic-disaster](https://github.com/JakeForsey/synthetic-disaster) -\u003e Generate synthetic satellite images of natural disasters using deep neural networks\n\n- [STPLS3D](https://github.com/meidachen/STPLS3D) -\u003e A Large-Scale Synthetic and Real Aerial Photogrammetry 3D Point Cloud Dataset\n\n- [LESS](https://github.com/jianboqi/lessrt) -\u003e LargE-Scale remote sensing data and image Simulation framework over heterogeneous 3D scenes\n\n- [Synthesizing Robustness: Dataset Size Requirements and Geographic Insights](https://avanetten.medium.com/synthesizing-robustness-dataset-size-requirements-and-geographic-insights-a687192e8004) -\u003e Medium article, concludes that synthetic data is most beneficial to the rarest object classes and that extracting utility from synthetic data often takes significant effort and creativity\n\n- [rs_img_synth](https://github.com/gbaier/rs_img_synth) -\u003e Synthesizing Optical and SAR Imagery From Land Cover Maps and Auxiliary Raster Data\n\n- [OnlyPlanes](https://github.com/naivelogic/OnlyPlanes) -\u003e dataset and pretrained models for the paper: OnlyPlanes - Incrementally Tuning Synthetic Training Datasets for Satellite Object Detection\n\n- [Using Stable Diffusion to Improve Image Segmentation Models](https://medium.com/edge-analytics/using-stable-diffusion-to-improve-image-segmentation-models-1e99c25acbf) -\u003e Augmenting Data with Stable Diffusion\n\n- [synthetic-satellite-imagery](https://github.com/ms-synthetic-satellite-image/synthetic-satellite-imagery) -\u003e Label-conditional Synthetic Satellite Imagery - generating synthetic satellite images and conducting downstream experiments\n\n#\n## Large vision \u0026 language models (LLMs \u0026 LVMs)\n\n- [awesome-remote-sensing-vision-language-models](https://github.com/lzw-lzw/awesome-remote-sensing-vision-language-models)\n\n- [Awesome-Remote-Sensing-Multimodal-Large-Language-Model](https://github.com/ZhanYang-nwpu/Awesome-Remote-Sensing-Multimodal-Large-Language-Model)\n\n- [Remote-Sensing-ChatGPT](https://github.com/HaonanGuo/Remote-Sensing-ChatGPT) -\u003e an open source tool for solving remote sensing tasks with ChatGPT in an interactive way.\n\n- [ChangeCLIP](https://github.com/dyzy41/ChangeCLIP) -\u003e ChangeCLIP: Remote sensing change detection with multimodal vision-language representation learning\n\n- [SkyEyeGPT](https://github.com/ZhanYang-nwpu/SkyEyeGPT) -\u003e SkyEyeGPT: Unifying Remote Sensing Vision-Language Tasks via Instruction Tuning with Large Language Model\n\n- [RemoteCLIP](https://github.com/ChenDelong1999/RemoteCLIP) -\u003e A Vision Language Foundation Model for Remote Sensing\n\n- [GeoChat](https://github.com/mbzuai-oryx/GeoChat) -\u003e Grounded Large Vision-Language Model for Remote Sensing\n\n- [labs-gpt-stac](https://github.com/developmentseed/labs-gpt-stac) -\u003e connect ChatGPT to a STAC API backend\n\n- [EarthGPT](https://github.com/wivizhang/EarthGPT) -\u003e A Universal Multi-modal Large Language Model for Multi-sensor Image Comprehension in Remote Sensing Domain\n\n- [H2RSVLM](https://github.com/opendatalab/H2RSVLM) -\u003e Towards Helpful and Honest Remote Sensing Large Vision Language Model\n\n- [LLMs \u0026 FMs in Smart Agriculture](https://arxiv.org/pdf/2308.06668) -\u003e Large Language Models and Foundation Models in Smart Agriculture: Basics, Opportunities, and Challenges\n\n- [LHRS-Bot](https://github.com/NJU-LHRS/LHRS-Bot) -\u003e Empowering Remote Sensing with VGI-Enhanced Large Multimodal Language Model\n\n- [Awesome-VLGFM](https://github.com/zytx121/Awesome-VLGFM) -\u003e Towards Vision-Language Geo-Foundation Models: A Survey\n\n#\n## Foundational models\n\n- [Awesome Remote Sensing Foundation Models](https://github.com/Jack-bo1220/Awesome-Remote-Sensing-Foundation-Models)\n\n- [Clay Foundation Model](https://github.com/Clay-foundation/model) -\u003e an open source AI model and interface for Earth.\n\n- [TerraTorch](https://github.com/IBM/terratorch) -\u003e a Python toolkit for fine-tuning Geospatial Foundation Models from IBM, based on PyTorch Lightning and TorchGeo\n\n- [EarthPT](https://github.com/aspiaspace/earthPT) -\u003e A time series foundation model for Earth Observation\n\n- [SpectralGPT](https://github.com/danfenghong/IEEE_TPAMI_SpectralGPT) -\u003e Spectral remote sensing foundation model, with finetuning on classification, segmentation, and change detection tasks\n\n- [DOFA-pytorch](https://github.com/zhu-xlab/DOFA) -\u003e Dynamic One-For-All (DOFA) multimodal foundation models for Earth vision reference implementation\n\n- [Prithvi foundation model](https://github.com/NASA-IMPACT/hls-foundation-os) -\u003e also see the [Baseline Model for Segmentation](https://github.com/ClarkCGA/multi-temporal-crop-classification-baseline)\n\n- [prithvi-pytorch](https://github.com/isaaccorley/prithvi-pytorch) -\u003e makes Prithvi usable from Pytorch Lightning\n\n- [geo-bench](https://github.com/ServiceNow/geo-bench) -\u003e a General Earth Observation benchmark for evaluating the performances of large pre-trained models on geospatial data\n\n- [USat](https://github.com/stanfordmlgroup/USat) -\u003e A Unified Self-Supervised Encoder for Multi-Sensor Satellite Imagery\n\n- [hydro-foundation-model](https://github.com/isaaccorley/hydro-foundation-model) -\u003e A Foundation Model for Water in Satellite Imagery\n\n- [RSBuilding](https://github.com/Meize0729/RSBuilding) -\u003e Towards General Remote Sensing Image Building Extraction and Change Detection with Foundation Model\n\n- [Text2Seg](https://github.com/Douglas2Code/Text2Seg) -\u003e  a pipeline that combined multiple Vision Foundation Models (SAM, CLIP, GroundingDINO) to perform semantic segmentation.\n\n- [Remote-Sensing-RVSA](https://github.com/ViTAE-Transformer/Remote-Sensing-RVSA) -\u003e Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model\n\n- [FoMo-Bench](https://github.com/RolnickLab/FoMo-Bench) -\u003e a multi-modal, multi-scale and multi-task Forest Monitoring Benchmark for remote sensing foundation models\n\n- [MTP](https://github.com/ViTAE-Transformer/MTP) -\u003e Advancing Remote Sensing Foundation Model via Multi-Task Pretraining\n\n- [DiffusionSat](https://www.samarkhanna.com/DiffusionSat/) -\u003e A Generative Foundation Model For Satellite Imagery\n\n- [granite-geospatial-biomass](https://github.com/ibm-granite/granite-geospatial-biomass) -\u003e A geospatial model for Above Ground Biomass from IBM\n\n- [RSP](https://github.com/ViTAE-Transformer/RSP) -\u003e An Empirical Study of Remote Sensing Pretraining\n\n- [geo-bench](https://github.com/ServiceNow/geo-bench) -\u003e GEO-Bench is a General Earth Observation benchmark for evaluating the performances of large pre-trained models on geospatial data. Read the full [paper](https://arxiv.org/abs/2306.03831)\n\n- [RS5M](https://github.com/om-ai-lab/RS5M) -\u003e RS5M and GeoRSCLIP: A Large Scale Vision-Language Dataset and A Vision-Language Foundation Model for Remote Sensing\n\n- [Galileo](https://github.com/nasaharvest/galileo) -\u003e Learning Global and Local Features in Pretrained Remote Sensing Models, from Nasa Harvest\n\n- [AnySat](https://github.com/gastruc/AnySat) -\u003e One Earth Observation Model for Many Resolutions, Scales, and Modalities\n\n----\n- *Logo created with* [*Brandmark*](https://app.brandmark.io/v3/)\n","funding_links":["https://github.com/sponsors/robmarkcole"],"categories":["Others","Other awesome awesome repositories","对象检测、分割","Uncategorized","Additional Resources"],"sub_categories":["Workshops","网络服务_其他","Uncategorized","AWESOME Resources List"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsatellite-image-deep-learning%2Ftechniques","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsatellite-image-deep-learning%2Ftechniques","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsatellite-image-deep-learning%2Ftechniques/lists"}