/SSFCN

[IEEE TBD 2020] Spectral-Spatial Fully Convolutional Networks for Hyperspectral Image Classification

Primary LanguagePythonMIT LicenseMIT

Spectral-Spatial Fully Convolutional Networks for Hyperspectral Image Classification

Tensorflow implementation of our method for hyperspectral image classification.

Paper

Beyond the Patchwise Classification: Spectral-Spatial Fully Convolutional Networks for Hyperspectral Image Classification

Please cite our papers if you find it useful for your research.

@article{ssfcn,
  title={Beyond the Patchwise Classification: Spectral-Spatial Fully Convolutional Networks for Hyperspectral Image Classification},
  author={Xu, Yonghao and Du, Bo and Zhang, Liangpei},
  journal={IEEE Transactions on Big Data},
  year={2020},
  volume={6},
  number={3},
  pages={492-506},
  publisher={IEEE}
}

Installation

  • Install Tensorflow 1.13 with Python 3.6.

  • Install pydensecrf.

  • Clone this repo.

git clone https://github.com/YonghaoXu/SSFCN

Dataset

Usage

  • For Spectral Classification:
    • Run SpeFCN.py
  • For Spatial Classification:
    • Run SpaFCN.py
  • For Spectral-Spatial Classification:
    • Run SSFCN.py