Main direction - An effective statistic for hyperspectral image classification

Manli Han, Yan Feng, Chao Xu, Jinazuo Sun, Ming Wang, Shaohui Mei

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

Abstract

In the hyperspectral image classification literature, mean and covariance are prominently used statistics. The first order statistics, i.e., mean, although easily estimated, can have low discriminative property. However, the discrimination power of the second order statistic, i.e., covariance, is strong. The downside of the estimation of covariance is the large sample size. In this paper, we propose a effective statistic named main direction. The main direction is defined as a direction in which the variation or the distance of the samples becomes the largest. Besides, the estimation of the main direction is stable even when the size of the sample set is very small. Based on the main direction, we propose a modified spectral angle mapper classifier and demonstrate the effectiveness of the modified classifier with experimental results.

Original languageEnglish
Title of host publicationICNC-FSKD 2018 - 14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery
EditorsZheng Xiao, Lipo Wang, Guoqing Xiao, Xiong Ning, Kenli Li, Maozhen Li
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages283-288
Number of pages6
ISBN (Electronic)9781538680971
DOIs
StatePublished - 2 Jul 2018
Event14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2018 - Huangshan, Anhui, China
Duration: 28 Jul 201830 Jul 2018

Publication series

NameICNC-FSKD 2018 - 14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery

Conference

Conference14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2018
Country/TerritoryChina
CityHuangshan, Anhui
Period28/07/1830/07/18

Keywords

  • Classification
  • Hyperspectral images
  • Main direction
  • Spectral angle mapper

Fingerprint

Dive into the research topics of 'Main direction - An effective statistic for hyperspectral image classification'. Together they form a unique fingerprint.

Cite this