xiuheng-wang / Tuning_free_PnP_HSI_deconvolution

Codes for the paper "Tuning-free Plug-and-Play Hyperspectral Image Deconvolution with Deep Priors" (T-GRS 2023).

Tuning-free Plug-and-Play Hyperspectral Image Deconvolution with Deep Priors

Steps for the public CAVE dataset:

1. Run cave_processing.py to process the data;

2. Run blurring_image.m to blurring the raw hyperspectral images;

3. Run main.sh to test our method.

For our real-world dataset (./data/Hide/):

Simply run main_real.sh to test our method.

The trained parameters of the B3DDN is stored in ./models/hsidb_epoch500.pkl, if you want to train the B3DDN by yourself:

1. Run cave_processing.py to process the data;

2. Run train.py.

For any questions, feel free to email me at [e-mail].

If this code is helpful for you, please cite our paper as follows:

@article{wang2023tuning,
  title={Tuning-free Plug-and-Play Hyperspectral Image Deconvolution with Deep Priors},
  author={Wang, Xiuheng and Chen, Jie and Richard, C{\'e}dric},
  journal={IEEE Transactions on Geoscience and Remote Sensing},
  year={2023}
}
@inproceedings{wang2020learning,
  title={Learning Spectral-Spatial Prior Via 3DDNCNN for Hyperspectral Image Deconvolution},
  author={Wang, Xiuheng and Chen, Jie and Richard, C{\'e}dric and Brie, David},
  booktitle={ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  pages={2403--2407},
  year={2020},
  organization={IEEE}
}