acecreamu / deep-hs-prior

Single Hyperspectral Image Denoising, Inpainting, Super-Resolution

Deep Hyperspectral Prior: Single Image Denoising, Inpainting, Super-Resolution

Supplementary code to the paper O Sidorov, JY Hardeberg. Deep Hyperspectral Prior: Denoising, Inpainting, Super-Resolution in ICCV 2019 Workshops.

Get started

The implementation is based on original Deep Image Prior code by Dmitry Ulyanov.

The framework was modified to process hyperspectral data using 2D or 3D convolutions:

Requirements

• python = 3.6
• pytorch = 0.4
• numpy
• scipy
• matplotlib
• scikit-image
• jupyter

Prepare the data

• Input and output hyperspectral data is contained in *.mat files.
• Specify a path to the file and name of the variable to read.
  For example, if data is contained in variable image:

  file_name  = 'data/inpainting/inpainting192.mat'
  mat = scipy.io.loadmat(file_name)
  img_np = mat["image"]
  

• Use custom code or one of the *.m files located at data/%task%/ to generate *.mat file.

Run the code

• Follow one of the proposed notebook files to get the results.
  * 2D versions tend to demonstrate better accuracy.
• Try to modify parameters. Have fun.

---

Some results

Denoising:

Super-Resolution:

Inpainting:

---

Please, kindly cite the paper if you use the code!

deep-hs-prior's Issues

How to use custom code or one of the *.m files located at data/%task%/ to generate *.mat file

I don't know how to run .m file in conda environment?
For example, the first line in /data/sr/prepare_sr.m:

what does pavia mean? I try to use matlab in my local laptop, while it cann't find pavia:

I would appreciate it if you can reply me as soon as possible. Thany youvery much.

[denoising2D] Error : No such file or directory: 'results/result_denoising_2D_it000

I was trying to run denoising2D file. Downloaded DC dataset, generated .mat file. In the last Block titled "Optimize" I get this error :

FileNotFoundError Traceback (most recent call last)
/usr/local/lib/python3.7/dist-packages/scipy/io/matlab/mio.py in _open_file(file_like, appendmat, mode)
38 try:
---> 39 return open(file_like, mode), True
40 except IOError:

FileNotFoundError: [Errno 2] No such file or directory: 'results/result_denoising_2D_it00000.mat'

During handling of the above exception, another exception occurred:

FileNotFoundError Traceback (most recent call last)
6 frames
/usr/local/lib/python3.7/dist-packages/scipy/io/matlab/mio.py in _open_file(file_like, appendmat, mode)
43 if appendmat and not file_like.endswith('.mat'):
44 file_like += '.mat'
---> 45 return open(file_like, mode), True
46 else:
47 raise IOError('Reader needs file name or open file-like object')

FileNotFoundError: [Errno 2] No such file or directory: 'results/result_denoising_2D_it00000.mat'

Due to this line,
if i % save_every == 0: scipy.io.savemat("results/result_denoising_2D_it%05d.mat" % (i), {'pred':out_np.transpose(1,2,0), 'pred_avg':out_avg_np.transpose(1,2,0)})
What should be done to resolve it?

请问prepare_denoisning.m文件中的I1 = double(dc);中dc是什么？

Missing data?

When I try to run "prepare_sr.m" I get an error that says "undefined function or variable pavia". The Matlab code starts without defining it.

the net output shape problem

Dear author,
Thank you so much for your code.
When I running the Optimize part in the "denoising.ipynb", having a problem as below:
RuntimeError: The size of tensor a (192) must match the size of tensor b (200) at non-singleton dimension 3

Then, I find the out size will become [1, 191, 192, 192] when through the "net", could you tell me how to solve this problem? Thank you very much!