Matlab Code for Noise-Robust Hyperspectral Image Classification via Multi-Scale Total Variation

Puhong Duan, Xudong Kang, Shutao Li, Pedram Ghamisi

---

The code in this toolbox implements the "Noise-Robust Hyperspectral Image Classification via Multi-Scale Total Variation". More specifically, it is detailed as follow.

Citation

Please kindly cite the papers if this code is useful and helpful for your research.

P. Duan, X. Kang, S. Li and P. Ghamisi, "Noise-Robust Hyperspectral Image Classification via Multi-Scale Total Variation," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 12, no. 6, pp. 1948-1962, June 2019, doi: 10.1109/JSTARS.2019.2915272.

@article{MSTV,
  title     = {Noise-Robust Hyperspectral Image Classification via Multi-Scale Total Variation},
  author    = {Duan, Puhong and Kang, Xudong and Li, Shutao and Ghamisi, Pedram},
  journal   = {IEEE Transactions on Geoscience and Remote Sensing}, 
  volume    = {12},
  number    = {6},
  pages     = {1948-1962},
  year      = {2019},
  publisher = {IEEE}
 }

Usage

run demo.m in Matlab

The SVM toolbox can be downloaded at: http://www.csie.ntu.edu.tw/~cjlin/libsvm/