Leopold-Lin / GMMDP_for_HIC

GMMDP, hyperspectral images classification

Multiview Marginal Discriminant Projection for Hyperspectral Images Classification

1. Introduction

​ This is the source code of our NCIG 2018 paper "Multiview Marginal Discriminant Projection for Hyperspectral Images Classification" (full-text)and JVCI paper "Graph regularized multiview marginal discriminant projection" (full-text).

​ We employed multiview subspace learning for feature reduction with the problems of high feature dimension and redundant information of hyperspectral images, and proposed a graph regularized multiview marginal discriminant projection (GMMDP) algorithm. The multiview feature reduction algorithm took the spectral features of each pixels as a view and spatial features as another view, then searched the optimal discriminant common subspace by optimizing the projection direction of each view.

2. Dependency

GMMDP is written by Python 3.6 and following libs are needed:

• sklearn
• numpy
• scipy
• pywt

3. Demo

workspace.py is the entrance of program.

print('MvDA', 'indian', 'wavelet')
for i in range(20):
    experiment('MvDA', 'indian', 'wavelet', 20, 0.45)

python3 workspace.py

4. Reference

Please cite the papers if you use our code.

• GB/T 7714

  Pan H, He J, Ling Y, et al. Graph Regularized Multiview Marginal Discriminant Projection[J]. Journal of Visual Communication and Image Representation.

• MLA

  Pan, Heng, et al. "Graph Regularized Multiview Marginal Discriminant Projection." Journal of Visual Communication and Image Representation.

• APA

  Pan, H., He, J., Ling, Y., Ju, L., & He, G. . Graph regularized multiview marginal discriminant projection. Journal of Visual Communication and Image Representation.

5. Contact

Contact me if you have any questions about the code and its execution.

[e-mail]