Houston datasets: Houston 2013 and Houston 2018. | Thanks Xudong Zhao for his work in producing the Houston data.
download: 坚果云 or Google-drive
Pavia datasets: University of Pavia (UP) and Pavia Center (PC).
download: 坚果云 or Google-drive
HyRank datasets: Dioni and Loukia.
download: 坚果云 or Google-drive
The dataset directory should look like this:
datasets
├── Houston
│ ├── Houston13.mat
│ ├── Houston13_7gt.mat
│ ├── Houston18.mat
│ └── Houston18_7gt.mat
├── Pavia
│ ├── paviaC.mat
│ └── paviaC_7gt.mat
│ ├── paviaU.mat
│ └── paviaU_7gt.mat
└── HyRANK
├── Dioni.mat
└── Dioni_gt_out68.mat
├── Loukia.mat
└── Loukia_gt_out68.mat- The variable names of data and gt in .mat file are set as
ori_dataandmap. - Pseudo-color image of all data are provided in
./figurefor use in writing papers. - The HyRANK dataset is screened for classes and samples. The gt used in the experiment is
*_gt_out68.mat.
Please cite our paper if you find these datasets useful for your research.
@ARTICLE{9540028,
author={Zhang, Yuxiang and Li, Wei and Zhang, Mengmeng and Qu, Ying and Tao, Ran and Qi, Hairong},
journal={IEEE Transactions on Neural Networks and Learning Systems},
title={Topological Structure and Semantic Information Transfer Network for Cross-Scene Hyperspectral Image Classification},
year={2021},
volume={},
number={},
pages={1-14},
doi={10.1109/TNNLS.2021.3109872}}
@ARTICLE{9812472,
author={Zhang, Yuxiang and Li, Wei and Zhang, Mengmeng and Wang, Shuai and Tao, Ran and Du, Qian},
journal={IEEE Transactions on Neural Networks and Learning Systems},
title={Graph Information Aggregation Cross-Domain Few-Shot Learning for Hyperspectral Image Classification},
year={2022},
volume={},
number={},
pages={1-14},
doi={10.1109/TNNLS.2022.3185795}}