Texture classification using hierarchical linear discriminant space

Yousun Kang, Kenichi Morooka, Hiroshi Nagahashi

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

As a representative of the linear discriminant analysis, the Fisher method is most widely used in practice and it is very effective in two-class classification. However, when it is expanded to a multi-class classification problem, the precision of its discrimination may become worse. A main reason is an occurrence of overlapped distributions on the discriminant space built by Fisher criterion. In order to take such overlaps among classes into consideration, our approach builds a new discriminant space by hierarchically classifying the overlapped classes. In this paper, we propose a new hierarchical discriminant analysis for texture classification. We divide the discriminant space into subspaces by recursively grouping the overlapped classes. In the experiment, texture images from many classes are classified based on the proposed method. We show the outstanding result compared with the conventional Fisher method.

Original languageEnglish
Pages (from-to)2380-2387
Number of pages8
JournalIEICE Transactions on Information and Systems
VolumeE88-D
Issue number10
DOIs
Publication statusPublished - 2005
Externally publishedYes

Fingerprint

Textures
Discriminant analysis
Experiments

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Texture classification using hierarchical linear discriminant space. / Kang, Yousun; Morooka, Kenichi; Nagahashi, Hiroshi.

In: IEICE Transactions on Information and Systems, Vol. E88-D, No. 10, 2005, p. 2380-2387.

Research output: Contribution to journalArticle

@article{2a0087e3f38b46029d4a48ee879b8ac1,
title = "Texture classification using hierarchical linear discriminant space",
abstract = "As a representative of the linear discriminant analysis, the Fisher method is most widely used in practice and it is very effective in two-class classification. However, when it is expanded to a multi-class classification problem, the precision of its discrimination may become worse. A main reason is an occurrence of overlapped distributions on the discriminant space built by Fisher criterion. In order to take such overlaps among classes into consideration, our approach builds a new discriminant space by hierarchically classifying the overlapped classes. In this paper, we propose a new hierarchical discriminant analysis for texture classification. We divide the discriminant space into subspaces by recursively grouping the overlapped classes. In the experiment, texture images from many classes are classified based on the proposed method. We show the outstanding result compared with the conventional Fisher method.",
author = "Yousun Kang and Kenichi Morooka and Hiroshi Nagahashi",
year = "2005",
doi = "10.1093/ietisy/e88-d.10.2380",
language = "English",
volume = "E88-D",
pages = "2380--2387",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "一般社団法人電子情報通信学会",
number = "10",

}

TY - JOUR

T1 - Texture classification using hierarchical linear discriminant space

AU - Kang, Yousun

AU - Morooka, Kenichi

AU - Nagahashi, Hiroshi

PY - 2005

Y1 - 2005

N2 - As a representative of the linear discriminant analysis, the Fisher method is most widely used in practice and it is very effective in two-class classification. However, when it is expanded to a multi-class classification problem, the precision of its discrimination may become worse. A main reason is an occurrence of overlapped distributions on the discriminant space built by Fisher criterion. In order to take such overlaps among classes into consideration, our approach builds a new discriminant space by hierarchically classifying the overlapped classes. In this paper, we propose a new hierarchical discriminant analysis for texture classification. We divide the discriminant space into subspaces by recursively grouping the overlapped classes. In the experiment, texture images from many classes are classified based on the proposed method. We show the outstanding result compared with the conventional Fisher method.

AB - As a representative of the linear discriminant analysis, the Fisher method is most widely used in practice and it is very effective in two-class classification. However, when it is expanded to a multi-class classification problem, the precision of its discrimination may become worse. A main reason is an occurrence of overlapped distributions on the discriminant space built by Fisher criterion. In order to take such overlaps among classes into consideration, our approach builds a new discriminant space by hierarchically classifying the overlapped classes. In this paper, we propose a new hierarchical discriminant analysis for texture classification. We divide the discriminant space into subspaces by recursively grouping the overlapped classes. In the experiment, texture images from many classes are classified based on the proposed method. We show the outstanding result compared with the conventional Fisher method.

UR - http://www.scopus.com/inward/record.url?scp=33645692373&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645692373&partnerID=8YFLogxK

U2 - 10.1093/ietisy/e88-d.10.2380

DO - 10.1093/ietisy/e88-d.10.2380

M3 - Article

AN - SCOPUS:33645692373

VL - E88-D

SP - 2380

EP - 2387

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 10

ER -