ISSLDIA

This repository contains the Matlab codes of paper "ISSLDIA: Internal Structure Self-learning and Low Dimensional Image Assistance for Single Hyperspectral Image Super-Resolution"

DOI: https://doi.org/10.1016/j.infrared.2025.105859

Prepare

1. Please read ./ompbox/readme.txt and ./ksvdbox/readme.txt first and follow the steps to setup the ompbox and ksvdbox function tools.
2. Put your RGB training images anywhere you like. (you need change the training path in 'Training.m' matfile.)

    ├──{your training dataset name}
    |     └── *.{png/jpg/tif/...}      
    |     └── *.{png/jpg/tif/...}
    |     └──  ...
    ----------------------------------------------

We provided two training data sets under ./UCMerced_LandUse/Images/ and ./CVPR08-SR/Data/ in .zip format, you can unzip them and put these images under their own folder.

3. Put testing hyperspectral data in '.mat' format under

    ├──Data
    |  └──{your testing dataset name}
    |  |  └── *.mat
    |  |  └── *.mat   
    |  |  └──  ...  (more than one '.mat' file)
    |  └──{your testing dataset name}
    |     └── *.mat   (only one '.mat' file)
    ----------------------------------------------

Train

Run ./Training.m

Test

Run ./go_run.m

Hyperspectral datasets

1. CAVE
2. Pavia University
3. Washington DC mall
4. Chikusei

Contact

If you have any question, please email: ren_jinhao@163.com