Skip to main navigation Skip to search Skip to main content

Locality-preserving discriminant analysis for hyperspectral image classification using local spatial information

  • Wei Li*
  • , Saurabh Prasad
  • , Zhen Ye
  • , James E. Fowler
  • , Minshan Cui
  • *Corresponding author for this work
  • Mississippi State University
  • University of Houston
  • Northwestern Polytechnical University Xian

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Locality-preserving projection as well as local Fisher discriminant analysis is applied for dimensionality reduction of hyperspectral imagery based on both spatial and spectral information. These techniques preserve the local geometric structure of hyperspectral data into a low-dimensional subspace wherein a Gaussian-mixture-model classifier is then considered. In the proposed classification system, local spatial information - which is expected to be more multimodal than strictly spectral features - is used. Results with experimental hyperspectral data demonstrate that this system outperforms traditional classification approaches.

Original languageEnglish
Title of host publicationIGARSS 2012 - 2012 IEEE International Geoscience and Remote Sensing Symposium
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4134-4137
Number of pages4
ISBN (Electronic)9781467311595
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012 - Munich, Germany
Duration: 22 Jul 201227 Jul 2012

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)
ISSN (Print)2153-6996
ISSN (Electronic)2153-7003

Conference

Conference32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012
Country/TerritoryGermany
CityMunich
Period22/07/1227/07/12

Keywords

  • Dimensionality reduction
  • hyperspectral data
  • linear discriminant analysis
  • pattern classification

Fingerprint

Dive into the research topics of 'Locality-preserving discriminant analysis for hyperspectral image classification using local spatial information'. Together they form a unique fingerprint.

Cite this