aicip / uDAS

Unsupervised spectral unmixing through an untied denoising autoencoder with sparsity (uDAS)

This is a demo code for 'uDAS: An untied denoising autoencoder with sparsity for spectral unmixing', the code is for research purpose only, all rights reserved. 

Contact information 
Ying Qu: [e-mail]


To run the code, please use 'train_auto_syn.m' in Matlab. 


Please cite the following two paper. 

Qu, Ying, and Hairong Qi. "uDAS: An untied denoising autoencoder with sparsity for spectral unmixing." IEEE Transactions on Geoscience and Remote Sensing 57.3 (2019): 1698-1712.

Qu, Ying, Rui Guo, and Hairong Qi. "Spectral unmixing through part-based non-negative constraint denoising autoencoder." 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). IEEE, 2017.


@article{qu2019udas,
  title={uDAS: An untied denoising autoencoder with sparsity for spectral unmixing},
  author={Qu, Ying and Qi, Hairong},
  journal={IEEE Transactions on Geoscience and Remote Sensing},
  volume={57},
  number={3},
  pages={1698--1712},
  year={2019},
  publisher={IEEE}
}


@article{qu2017spectral,
  title={Spectral unmixing through part-based non-negative constraint denoising autoencoder},
  author={Qu, Ying and Guo, Rui and Qi, Hairong},
  journal ={2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)},
  pages={209--212},
  year={2017},
  organization={IEEE}
}