Skip to content

DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery, TCSVT, 2022

License

Notifications You must be signed in to change notification settings

B-Xi/TCSVT_2022_DGSSC

Repository files navigation

DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery, TCSVT, 2022

Bobo Xi, Jiaojiao Li, Yan Diao, Yunsong Li, Zan Li, Yan Huang, and Jocelyn Chanussot.


Code for the paper: DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery.

Fig. 1: Architecture of the proposed DGSSC for imbalanced hyperspectral imagery. The encoder, decoder, and classifier are trained in an end-to-end fashion, driven by the integrated loss of LPdRec, Lmmd, and Lcls.

Training and Test Process

Please run the 'main.py' to reproduce the DGSSC results on Lokia dataset.

We have successfully tested it on Ubuntu 16.04 with Tensorflow 1.13.1. Below is the classification result when 1% labeled samples are selected as the training set.

Fig. 2: The composite false-color image, groundtruth, and classification map of Lokia dataset.

References

If you find this code helpful, please kindly cite:

[1] B. Xi, J. Li, Y. Diao, Y. Li, Z. Li, Y. Huang, and J. Chanussot, "DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery," in IEEE Transactions on Circuits and Systems for Video Technology, pp. 1-1, 2022, doi: 10.1109/TCSVT.2022.3215513.

Citation Details

BibTeX entry:

@ARTICLE{Xi2022_TCSVT_DGSSC,
  author={Xi, Bobo and Li, Jiaojiao and Diao, Yan and Li, Yunsong and Li, Zan and Huang, Yan and Chanussot, Jocelyn},
  journal={IEEE Transactions on Circuits and Systems for Video Technology}, 
  title={DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery}, 
  year={2022},
  volume={},
  number={},
  pages={1-1},
  doi={10.1109/TCSVT.2022.3215513}}

Licensing

Copyright (C) 2022 Bobo Xi

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, version 3 of the License.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program.

About

DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery, TCSVT, 2022

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages