https://doi.org/10.5281/zenodo.14391084

The uploaded data contains two ZIP files: one is the original hyperspectral image (original_data_v2.zip), and the other is the hyperspectral image processed by VCA (data_v2.zip).

The data used for training is from data_v2.zip, which is the pre-processed file. Note: The preprocessing files are merely the results of preprocessing before training and should not be considered as the files to be compressed. This process can be performed in memory; however, for convenience in demonstration, the results are saved as MAT files.

Description: The uncompressed file. 

Description: The pre-processed file.

Data was derived from the following sources:

• Chikusei Dataset: https://naotoyokoya.com/Download.html
• Matiwan Dataset: https://aistudio.baidu.com/datasetdetail/100218
• DFC2018 Houston Dataset: http://dase.grss-ieee.org
• Pavia Centre, and Cuprite Dataset: Dataset: https://www.ehu.eus/ccwintco/index.php/Hyperspectral_Remote_Sensing_Scenes
• Indian Pines Dataset: https://purr.purdue.edu/publications/1947/1
• Jasper Ridge Dataset: https://lesun.weebly.com/hyperspectral-data-set.html