https://github.com/z-fran/satellite-image-classification
Train AI models on satellite image dataset to classify different types of land.
https://github.com/z-fran/satellite-image-classification
satellite-image-classification satellite-imagery-analysis
Last synced: 2 months ago
JSON representation
Train AI models on satellite image dataset to classify different types of land.
- Host: GitHub
- URL: https://github.com/z-fran/satellite-image-classification
- Owner: Z-Fran
- Created: 2024-08-09T07:17:21.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-08-09T15:08:49.000Z (10 months ago)
- Last Synced: 2025-02-08T13:42:56.495Z (4 months ago)
- Topics: satellite-image-classification, satellite-imagery-analysis
- Language: Python
- Homepage:
- Size: 4.83 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Satellite-Image-Classification
## Introduction
Train AI models on satellite image dataset to classify different types of land.
## How to run
Run [`train.ipynb`](./train.ipynb) to train models on satellite image dataset.
## Dataset
Segment satellite imagery into small blocks, and annotate type labels of blocks.
### Demo dataset
We provide a small [dataset](./dataset.zip) in the repository. You can directly unzip it and see the following folder structure.
```
Satellite-Image-Classification/
├── dataset
| ├── industry
| | ├── xxx.jpg
| | └── ...
| ├── agriculture
| └── residence
```
### Create your own dataset
1. Get a satellite imagery in `.tiff` format.
2. Use [Global Mapper](https://www.bluemarblegeo.com/global-mapper/) to split the `tiff` file into small blocks in `.jpg` format.
3. Manually classify these images and create a folder structure like above.
## Models
Models are listed in the folder `models`, which is imported from [pytorch-cifar
](https://github.com/kuangliu/pytorch-cifar).
## References
- [pytorch-cifar
](https://github.com/kuangliu/pytorch-cifar)