Advanced search
1 file | 361.63 KB

Feature extraction for hyperspectral images based on semi-supervised local linear discriminant analysis

Author
Organization
Project
FWO
Abstract
We propose a novel semi-supervised local discriminant analysis (SELD) method for feature extraction in hyperspectral remote sensing imagery. The proposed method combines a supervised method (Linear Discriminant Analysis (LDA)) and an unsupervised method (Neighborhood Preserving Embedding (NPE)) without any free parameters. The underlying idea is to design optimal projection vectors, which can discover the global discriminant structure of the available labeled samples while preserving the local neighborhood spatial structure of the unlabeled samples. Furthermore, in our approach the number of extracted feature bands is no longer limited by the number of classes, which is a disadvantage of LDA. Experimental results demonstrate that the proposed method outperforms consistently other related semi-supervised methods and that it is also much more stable when the percentage of the labeled samples changes.
Keywords
feature extraction, Hyperspectral remote sensing, classification

Downloads

    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 361.63 KB

Citation

Please use this url to cite or link to this publication:

Chicago
Liao, Wenzhi, Aleksandra Pizurica, Wilfried Philips, and Youguo Pi. 2011. “Feature Extraction for Hyperspectral Images Based on Semi-supervised Local Linear Discriminant Analysis.” In Joint Urban Remote Sensing Event (JURSE - 2011), Proceedings, ed. Uwe Stilla, P Gamba, C Juergens, and D Maktav, 401–404. IEEE.
APA
Liao, Wenzhi, Pizurica, A., Philips, W., & Pi, Y. (2011). Feature extraction for hyperspectral images based on semi-supervised local linear discriminant analysis. In Uwe Stilla, P. Gamba, C. Juergens, & D. Maktav (Eds.), Joint Urban Remote Sensing Event (JURSE - 2011), Proceedings (pp. 401–404). Presented at the Joint Urban Remote Sensing Event (JURSE - 2011), IEEE.
Vancouver
1.
Liao W, Pizurica A, Philips W, Pi Y. Feature extraction for hyperspectral images based on semi-supervised local linear discriminant analysis. In: Stilla U, Gamba P, Juergens C, Maktav D, editors. Joint Urban Remote Sensing Event (JURSE - 2011), Proceedings. IEEE; 2011. p. 401–4.
MLA
Liao, Wenzhi, Aleksandra Pizurica, Wilfried Philips, et al. “Feature Extraction for Hyperspectral Images Based on Semi-supervised Local Linear Discriminant Analysis.” Joint Urban Remote Sensing Event (JURSE - 2011), Proceedings. Ed. Uwe Stilla et al. IEEE, 2011. 401–404. Print.
@inproceedings{1216543,
  abstract     = {We propose a novel semi-supervised local discriminant analysis (SELD) method for feature extraction in hyperspectral remote sensing imagery. The proposed method combines a supervised method (Linear Discriminant Analysis (LDA)) and an unsupervised method (Neighborhood Preserving Embedding (NPE)) without any free parameters. The underlying idea is to design optimal projection vectors, which can discover the global discriminant structure of the available labeled samples while preserving the local neighborhood spatial structure of the unlabeled samples. Furthermore, in our approach the number of extracted feature bands is no longer limited by the number of classes, which is a disadvantage  of LDA. Experimental  results demonstrate that the proposed method outperforms consistently other related semi-supervised methods and that it is also much more stable when the percentage of the labeled samples changes.},
  author       = {Liao, Wenzhi and Pizurica, Aleksandra and Philips, Wilfried and Pi, Youguo},
  booktitle    = {Joint Urban Remote Sensing Event (JURSE - 2011), Proceedings},
  editor       = {Stilla, Uwe  and Gamba, P and Juergens, C and Maktav, D},
  isbn         = {9781424486571},
  keyword      = {feature extraction,Hyperspectral remote sensing,classification},
  language     = {eng},
  location     = {Munich, Germany},
  pages        = {401--404},
  publisher    = {IEEE},
  title        = {Feature extraction for hyperspectral images based on semi-supervised local linear discriminant analysis},
  url          = {http://archive.ugent.be/person/000091851926},
  year         = {2011},
}

Web of Science
Times cited: