{"id":13498967,"url":"https://github.com/gokriznastic/HybridSN","last_synced_at":"2025-03-29T03:32:18.609Z","repository":{"id":52494852,"uuid":"167841010","full_name":"gokriznastic/HybridSN","owner":"gokriznastic","description":"A keras based implementation of Hybrid-Spectral-Net as in IEEE GRSL paper \"HybridSN: Exploring 3D-2D CNN Feature Hierarchy for Hyperspectral Image Classification\".","archived":false,"fork":false,"pushed_at":"2023-12-23T07:24:11.000Z","size":70658,"stargazers_count":311,"open_issues_count":4,"forks_count":116,"subscribers_count":6,"default_branch":"master","last_synced_at":"2024-10-31T17:39:03.224Z","etag":null,"topics":["3d-cnn","hyperspectral-image-classification","hyperspectral-imaging","remote-sensing"],"latest_commit_sha":null,"homepage":"https://ieeexplore.ieee.org/document/8736016","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/gokriznastic.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"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}},"created_at":"2019-01-27T18:07:24.000Z","updated_at":"2024-10-27T12:30:23.000Z","dependencies_parsed_at":"2023-01-23T14:00:28.036Z","dependency_job_id":"c58e9cca-ce01-4fdb-9572-80c4f8f5f000","html_url":"https://github.com/gokriznastic/HybridSN","commit_stats":{"total_commits":67,"total_committers":2,"mean_commits":33.5,"dds":0.4328358208955224,"last_synced_commit":"8e9fd37cdcbff5f9717433cddb83f5215f9ec54e"},"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gokriznastic%2FHybridSN","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gokriznastic%2FHybridSN/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gokriznastic%2FHybridSN/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gokriznastic%2FHybridSN/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/gokriznastic","download_url":"https://codeload.github.com/gokriznastic/HybridSN/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246135767,"owners_count":20729056,"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":["3d-cnn","hyperspectral-image-classification","hyperspectral-imaging","remote-sensing"],"created_at":"2024-07-31T22:00:23.630Z","updated_at":"2025-03-29T03:32:15.599Z","avatar_url":"https://github.com/gokriznastic.png","language":"Jupyter Notebook","funding_links":[],"categories":["3 Code"],"sub_categories":["3.1 Comparison methods of our proposed EMS-GCN methods"],"readme":"# Hybrid-Spectral-Net for Hyperspectral Image Classification.\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/hybridsn-exploring-3d-2d-cnn-feature/hyperspectral-image-classification-on-indian)](https://paperswithcode.com/sota/hyperspectral-image-classification-on-indian?p=hybridsn-exploring-3d-2d-cnn-feature)\n\n## PyTorch Implimentation of HybridSN\n\nPyTorch version of the HybridSN is available: https://github.com/Pancakerr/HybridSN\n\n## Description\n\nThe  HybridSN  is  spectral-spatial  3D-CNN  followed  by spatial 2D-CNN. The 3D-CNN facilitates the joint spatial-spectral feature  representation  from  a  stack  of  spectral  bands.  The  2D-CNN  on  top  of  the  3D-CNN  further  learns  more  abstract  level spatial  representation. \n\n## Model\n\n\u003cimg src=\"figure/HSI-RN.jpg\"/\u003e\n\nFig: Proposed HybridSpectralNet (HybridSN) Model with 3D and 2D convolutions for hyperspectral image (HSI) classification.\n\n## Prerequisites\n\n- [Anaconda 2.7](https://www.anaconda.com/download/#linux)\n- [Tensorflow 1.3](https://github.com/tensorflow/tensorflow/tree/r1.3)\n- [Keras 2.0](https://github.com/fchollet/keras)\n\n## Results\n\n### Indian Pines (IP) dataset\n\n\u003cimg src=\"figure/IP-FC.jpg\" width=\"200\" height=\"200\"/\u003e \u003cimg src=\"figure/IP-GT.jpg\" width=\"200\" height=\"200\"/\u003e \u003cimg src=\"figure/IP-Pr.jpg\" width=\"200\" height=\"200\"/\u003e \u003cimg src=\"figure/IP_legend.jpg\" width=\"250\" height=\"150\"/\u003e\n\nFig.2  The IN dataset classification result (Overall Accuracy 99.81%) of Hybrid-SN using 30% samples for training. (a) False color image. (b) Ground truth labels. (c) Classification map. (d) Class legend. \n\n### University of Pavia (UP) dataset\n\n\u003cimg src=\"figure/UP-FC.jpg\"/\u003e \u003cimg src=\"figure/UP-GT.jpg\"/\u003e \u003cimg src=\"figure/UP-Pr.jpg\"/\u003e \u003cimg src=\"figure/UP_legend.jpg\" width=\"200\" height=\"100\"/\u003e\n\nFig.3  The UP dataset classification result (Overall Accuracy 99.99%) of Hybrid-SN using 30% samples for training. (a) False color image. (b) Ground truth labels. (c) Classification map. (d) Class legend.\n\n### Salinas Scene (SS) dataset\n\n\u003cimg src=\"figure/SA-FC.jpg\"/\u003e \u003cimg src=\"figure/SA-GT.jpg\"/\u003e \u003cimg src=\"figure/SA-Pr.jpg\"/\u003e \u003cimg src=\"figure/SA_legend.jpg\" width=\"300\" height=\"150\"/\u003e\n\nFig.4  The UP dataset classification result (Overall Accuracy 100%) of Hybrid-SN using 30% samples for training. (a) False color image. (b) Ground truth labels. (c) Classification map.\n\n#### Detailed results can be found in the [Supplementary Material](supplementary-material.pdf)\n\n## Citation\n\nIf you use [HybridSN](https://github.com/gokriznastic/HybridSN) and [A2S2K-ResNet](https://github.com/suvojit-0x55aa/A2S2K-ResNet) and [HSI-Survey](https://github.com/AnkurDeria/HSI-Traditional-to-Deep-Models) code in your research, we would appreciate a citation to both the original paper:\n\n\t@article{roy2019hybridsn,\n        \ttitle={HybridSN: Exploring 3D-2D CNN Feature Hierarchy for Hyperspectral Image Classification},\n\t\tauthor={Roy, Swalpa Kumar and Krishna, Gopal and Dubey, Shiv Ram and Chaudhuri, Bidyut B},\n\t\tjournal={IEEE Geoscience and Remote Sensing Letters},\n\t\tvolume={17},\n\t\tno.={2},\n\t\tpp.={277-281},\n\t\tyear={2020}\n\t\t}\n\t@article{roy2020attention,\n\t\ttitle={Attention-based adaptive spectral-spatial kernel resnet for hyperspectral image classification},\n\t\tauthor={Swalpa Kumar Roy, and Suvojit Manna, and Tiecheng Song, and Lorenzo Bruzzone},\n\t\tjournal={IEEE Transactions on Geoscience and Remote Sensing},\n\t\tvolume={59},\n\t\tno.={9},\n\t\tpp.={7831-7843},\n\t\tyear={2021},\n\t\tpublisher={IEEE}\n\t\t}\t\n\t@article{ahmad2021hyperspectral,\n  \t\ttitle={Hyperspectral Image Classification--Traditional to Deep Models: A Survey for Future Prospects},\n  \t\tauthor={Muhammad Ahmad, and Sidrah Shabbir, and Swalpa Kumar Roy, and Danfeng Hong, and Xin Wu, and Jing Yao, and Adil Mehmood Khan,\n\t\tand Manuel Mazzara, and Salvatore Distefano, and Jocelyn Chanussot},\n  \t\tjournal={IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},\n  \t\tyear={2022},\n  \t\tvolume={15},\n  \t\tpages={968-999},\n  \t\tdoi={10.1109/JSTARS.2021.3133021},\n  \t\tpublisher={IEEE}\n\t\t}\n   \n\n## Acknowledgement\n\nPart of this code is from a implementation of Classification of HSI using CNN by [Konstantinos Fokeas](https://github.com/KonstantinosF/Classification-of-Hyperspectral-Image).\n\n## License\n\nCopyright (c) 2019 Gopal Krishna. Released under the MIT License. See [LICENSE](LICENSE) for details.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgokriznastic%2FHybridSN","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgokriznastic%2FHybridSN","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgokriznastic%2FHybridSN/lists"}