/SSAN

Spectral-Spatial Attention Network for Hyperspectral Image Classification

Primary LanguagePythonMIT LicenseMIT

Spectral-Spatial Attention Network for Hyperspectral Image Classification

Paper

Spectral-Spatial Attention Networks for Hyperspectral Image Classification

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

@Article{ssan,
  AUTHOR = {Mei, Xiaoguang and Pan, Erting and Ma, Yong and Dai, Xiaobing and Huang, Jun  and Fan, Fan and Du, Qinglei and Zheng, Hong and Ma, Jiayi},
  TITLE = {Spectral-Spatial Attention Networks for Hyperspectral Image Classification},
  JOURNAL = {Remote Sensing},
  VOLUME = {11},
  YEAR = {2019},
  NUMBER = {8},
  ARTICLE-NUMBER = {963},
  URL = {http://www.mdpi.com/2072-4292/11/8/963},
  ISSN = {2072-4292},
  DOI = {10.3390/rs11080963}
}

Installation

  • Install Tensorflow 1.9.0 with Python 3.6.

  • Clone this repo

    git clone https://github.com/EtPan/SSAN

Dataset

Download the Pavia Center dataset and its corresponding ground-truth map.

Usage

1. Change the file path

​ Replace the file path for the hyperspectral data in save_indices.py and indices.py;Replace the file path for the ckpt files of the model in ssan.py and ssan_test.py.

2. Split the dataset

​ Run save_indices.py .

3. Training

​ Run ssan.py.

4. Testing and Evaluation

​ Run ssan_test.py.