BTCDNet: Bayesian Tile Attention Network for Hyperspectral Image Change Detection

---

Introduction

Official implementation of BTCDNet by Junshen Luo, Jiahe Li, Xinlin Chu, Sai Yang, Lingjun Tao and Qian Shi.

---

---

How to use it?

1. Installation

git clone https://github.com/JeasunLok/BTCDNet.git && cd BTCDNet
conda create -n BTCDNet python=3.9
conda activate BTCDNet
pip install -r requirements.txt

2. Download our datasets

Download our datasets then place them in data folder:

Baiduyun: https://pan.baidu.com/s/1hyye2fVxoUaOJ6YR_RUSJg (access code: js66)

Google Drive: https://drive.google.com/drive/folders/1xe9i95_noh8dBW7sGn_z4uIfePBsa8r2

3. Quick start to use our SOTA model BTCDNet

You should change the settings in config.yaml especially HSI_data_path and model_type then:

chmod +x demo.sh
./demo.sh

4. More detailed information

see config.yaml

---

Citation

Please kindly cite the papers if this code is useful and helpful for your research.

J. Luo, J. Li, X. Chu, S. Yang, L. Tao and Q. Shi, "BTCDNet: Bayesian Tile Attention Network for Hyperspectral Image Change Detection," in IEEE Geoscience and Remote Sensing Letters, vol. 22, pp. 1-5, 2025, Art no. 5504205, doi: 10.1109/LGRS.2025.3563897.

@article{luo2025btcdnet,
  title={BTCDNet: Bayesian Tile Attention Network for Hyperspectral Image Change Detection},
  author={Luo, Junshen and Li, Jiahe and Chu, Xinlin and Yang, Sai and Tao, Lingjun and Shi, Qian},
  journal={IEEE Geoscience and Remote Sensing Letters},
  year={2025},
  publisher={IEEE}
}

---

Contact Information

Junshen Luo: luojsh7@mail2.sysu.edu.cn

Junshen Luo is with School of Geography and Planning, Sun Yat-sen University, Guangzhou 510275, China

---