RgB2CAVE dataset for traning and test

RgB2CAVE dataset comes from the CAVE data set which comprises 32 scenes with a size of 512×512 and are captured by a cooled CCD camera named Apogee Alta U260 and cover the spectral range from 400 nm to 700 nm with a 10 nm spectral resolution containing 31 bands. Moreover, the RGB images covering the same scene as hyperspectral data are also available. To verify the model performance of solving generalized spectral super-resolution, we propose a dataset by spatially downsampling the Green channel of the original RGB image with a ratio of 1∕2.

The file named "CAVE_train.h5" contains training samples where the high-resolution Red and Blue channel is with size of 128×128 while the size of Green channel is 64×64. Note that, data augmentation has also been implemented in RgB2CAVE to enlarge the dataset eightfold. So there are 5200 samples. Moreover, we also give some test images in the "CAVE_test.mat". Details about generalized spectral super-resolution can be found in the following paper.

Paper: J. He, Q. Yuan, J. Li, and L. Zhang, "PoNet: A universal physical optimization-based spectral super-resolution network for arbitrary multispectral images," Information Fusion, vol. 80, pp. 205–225, 2022.

More information about the author can be found at https://jianghe96.github.io/

If this dataset is helpful please cite as:

@article{He2022PoNet,
title={PoNet: A universal physical optimization-based spectral super-resolution network for arbitrary multispectral images},
author={He, Jiang and Yuan, Qiangqiang and Li, Jie and Zhang, Liangpei},
journal={Information Fusion},
volume={80},
pages={205--225},
year={2022},
}