Local discriminant non-negative matrix factorization feature extraction for hyperspectral image classification

Jing Wen, Yongqiang Zhao, Xiwen Zhang, Weidong Yan, Wei LinInternational Journal of Remote Sensing 2014 · DOI

Abstract

Non-negative matrix factorization (NMF) ignores both the local geometric structure of and the discriminative information contained in a data set. A manifold geometry-based NMF dimension reduction method called local discriminant NMF (LDNMF) is proposed in this paper. LDNMF preserves not only the non-negativity but also the local geometric structure and discriminative information of the data. The local geometric and discriminant structure of the data manifold can be characterized by a within-class graph and a between-class graph. An efficient multiplicative updating procedure is produced, and its global convergence is guaranteed theoretically. Experimental results on two hyperspectral image data sets show that the proposed LDNMF is a powerful and promising tool for extracting hyperspectral image features.
BibTeX
@article{wen2014local,
  author = {Jing Wen and Yongqiang Zhao and Xiwen Zhang and Weidong Yan and Wei Lin},
  title = {{Local discriminant non-negative matrix factorization feature extraction for hyperspectral image classification}},
  journal = {International Journal of Remote Sensing},
  year = {2014},
  doi = {10.1080/01431161.2014.930198},
}

Loading related content...

Figure