INTERPRETABLE MULTIPLE LOSS FUNCTIONS IN A LOW-RANK DEEP IMAGE PRIOR BASED METHOD FOR SINGLE HYPERSPECTRAL IMAGE SUPER-RESOLUTION

This repository provides the Python source codes related to the conference "INTERPRETABLE MULTIPLE LOSS FUNCTIONS IN A LOW-RANK DEEP IMAGE PRIOR BASED METHOD FOR SINGLE HYPERSPECTRAL IMAGE SUPER-RESOLUTION" presented in EUSIPCO 2021.

Installation

List of libraries required to execute the code.:

• python = 3.7.7
• Tensorflow = 2.2
• Keras = 2.4.3
• numpy
• scipy
• matplotlib
• h5py = 2.10
• opencv = 4.10
• poppy = 0.91

All of them can be installed via conda (anaconda), e.g.

conda install jupyter

or using pip install and the required file.

Data

This work uses the following three datasets. Please download the datasets and store them it correctly in the corresponding dataset folder:

• Toy from CAVE dataset: Provided in the Data_set/ folder. Available in https://www.cs.columbia.edu/CAVE/databases/multispectral/. Accessed: 20-Nov-2020
• Pavia dataset: Provided in the Data_set/ folder. Available in http://www.ehu.eus/ccwintco/index.php?title=Hyperspectral_Remote_Sensing_Scenes\#Pavia_Centre_and_University. Accessed: 22-Oct-2020

Structure of directories

Directory	Description
Data_set	Folder that contains the datasets.