Hyperspectral Remote Sensing Images Salient Object Detection: The First Benchmark Dataset and Baseline [TGRS 2025]

July 29, 2025 · View on GitHub

by Peifu Liu, Huiyan Bai, Tingfa Xu, Jihui Wang, Huan Chen, and Jianan Li.

arXiv

Requirements

It is recommended to use Python 3.9. GDAL is available at Google Drive. Feel free to use any PyTorch version.

Dataset

The HRSSD is available at Baidu Netdisk and Google Drive. Please place the dataset in the dataset folder. Our HRSSD is organized as follows:

/HRSSD
    /tr
        /image
        /label
        /mask
    /ts
        /image
        /label
        /mask

Compared Methods

  1. MJRBM: Paper, Code
  2. FSMINet: Paper, Code
  3. CorrNet: Paper, Code
  4. ACCoNet: Paper, Code
  5. SeaNet: Paper, Code
  6. MEANet: Paper, Code
  7. CTDNet: Paper, Code
  8. TRACER: Paper, Code
  9. BBRF: Paper, Code
  10. MENet: Paper, Code
  11. ADMNet: Paper, Code
  12. SED/SG: Paper, Code
  13. SUDF: Paper, Code
  14. SMN: Paper, Code
  15. CSCN: Paper, Code
  16. SAHRNet: Paper, Code
  17. MambaHSI: Paper, Code
  18. MambaLG: Paper, Code
  19. DSTC: Paper, Code

Acknowledgement

Our dataset is built upon WHU-OHS. Thanks for their great work!

License

This repository contains two components with different licenses:

Our code is released under the MIT License.

The HRSSD dataset is licensed under License: CC BY-NC-ND 4.0. It is intended for academic research only. You must attribute the original source, and you are not allowed to modify or redistribute the dataset without permission.