Skip to content

Codes for the paper "Hyperspectral super-resolution by unsupervised convolutional neural network and SURE" in Proceeding of IGARSS 2022, Kuala Lumpur, Malaysia.

Notifications You must be signed in to change notification settings

hvn2/SURE-MS-HS

Folders and files

NameName
Last commit message
Last commit date

Latest commit

c781624 · Oct 19, 2022

History

11 Commits
May 24, 2022
May 24, 2022
May 24, 2022
May 24, 2022
May 24, 2022
May 24, 2022
Oct 19, 2022
May 24, 2022

Repository files navigation

SURE-MS-HS

Offical Python codes for the paper "Hyperspectral super-resolution by unsupervised convolutional neural network and SURE", in Proceeding of IGARSS 2022, Kuala Lumpur, July 2022.

Authors: Han V. Nguyen , Magnus O. Ulfarsson , Johannes R. Sveinsson , and Mauro Dalla Mura
Faculty of Electrical and Computer Engineering, University of Iceland, Reykjavik, Iceland
Department of Electrical and Electronic Engineering, Nha Trang University, Khanh Hoa, Vietnam
GIPSA-Lab, Grenoble Institute of Technology, Saint Martin d’Hères, France.
Email: [email protected]

Abstract:

Recent advances in deep learning (DL) reveal that the structure of a convolutional neural network (CNN) is a good image prior (called deep image prior (DIP)), bridging the model-based and DL-based methods in image restoration. However, optimizing a DIP-based CNN is prone to overfitting leading to a poorly reconstructed image. This paper derives a loss function based on Stein's unbiased risk estimate (SURE) for unsupervised training of a DIP-based CNN applied to the hyperspectral image (HSI) super-resolution. The SURE loss function is an unbiased estimate of the mean-square-error (MSE) between the clean low-resolution image and the low-resolution estimated image, which relies only on the observed low-resolution image. Experimental results on HSI show that the proposed method not only improves the performance, but also avoids overfitting.

Please cite our paper if you are interested
@inproceedings{nguyen2022hyperspectral, title={Hyperspectral Super-Resolution by Unsupervised Convolutional Neural Network and Sure}, author={Nguyen, Han V and Ulfarsson, Magnus O and Sveinsson, Johannes R and Dalla Mura, Mauro}, booktitle={IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium}, pages={903--906}, year={2022}, organization={IEEE} }

Usage:

The following folders contanin:

  • data: The simulated PU dataset.
  • models: python scripts define the model (network structure)
  • utils: additional functions
    Run the jupyter notebook and see results.

Environment

  • Pytorch 1.8
  • Matplotlib
  • Numpy, Scipy, Skimage.

About

Codes for the paper "Hyperspectral super-resolution by unsupervised convolutional neural network and SURE" in Proceeding of IGARSS 2022, Kuala Lumpur, Malaysia.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published