- Download datasets from (new link) https://drive.google.com/file/d/0B1kqE05deSqxYm8wS1F3UXZYeWM/view?usp=drivesdk&resourcekey=0-NGlaHn82mfJaM9QNXb-cKA
- Extract datasets.zip to project directory.
- Run demo.m to visualize results for a selected dataset.
Code will be released soon on this page. Stay tuned!...
Samson Data and Abundance Map Results
Jasper Data and Abundance Map Results
Urban Data and Abundance Map Results
Cuprite Data and Abundance Map Results
Pavia University Data and Abundance Map Results
Pavia Center Data and Abundance Map Results
Gulfport Data and Abundance Map Results
IEEE 2013 Challenge Data and Abundance Map Results
DC Data and Abundance Map Results
Please cite the following paper:
[1] Savas Ozkan, Berk Kaya and Gozde Bozdagi Akar, EndNet: Sparse AutoEncoder Network for Endmember Extraction and Hyperspectral Unmixing, IEEE Transactions on Geoscience and Remote Sensing, 2018:
@article{ozkan2018endnet,
title={Endnet: Sparse autoencoder network for endmember extraction and hyperspectral unmixing},
author={Ozkan, Savas and Kaya, Berk and Akar, Gozde Bozdagi},
journal={IEEE Transactions on Geoscience and Remote Sensing},
volume={57},
number={1},
pages={482--496},
year={2018},
publisher={IEEE}
}
[2] Savas Ozkan and Gozde Bozdagi Akar, Spectral Unmixing With Multinomial Mixture Kernel and Wasserstein Generative Adversarial Loss, Advances in Neural Information Processing Systems Workshops, 2020:
@article{ozkan2020spectral,
title={Spectral Unmixing With Multinomial Mixture Kernel and Wasserstein Generative Adversarial Loss},
author={Ozkan, Savas and Akar, Gozde Bozdagi},
journal={arXiv preprint arXiv:2012.06859},
year={2020}
}