Discriminant eigenfaces: A new ranking method for principal components analysis

Página do item simplificado
dc.contributor.authorCarlos E. Thomaz
dc.contributor.authorGiraldi G. A.
dc.contributor.authorOrcidhttps://orcid.org/0000-0001-5566-1963
dc.date.accessioned2022-01-12T22:05:05Z
dc.date.available2022-01-12T22:05:05Z
dc.date.issued2008-10-26
dc.description.abstractPrincipal Component Analysis (PCA) is one of the most successful approaches to the problem of creating a low dimensional data representation and interpretation. However, since PCA explains the covariance structure of all the data, the first principal components with the largest eigenvalues do not necessarily represent important discriminant directions to separate sample groups. In this work, we investigate a new ranking method for the principal components. Instead of sorting the principal components in decreasing order of the corresponding eigenvalues, we propose the idea of using the discriminant weights given by separating hyperplanes to select among the principal components the most discriminant ones. Our experimental results have shown that the principal components selected by the separating hyperplanes are quite useful for understanding the differences between sample groups in face image analysis, allowing robust reconstruction and interpretation of the data as well as higher recognition rates using less linear features. © 2008 Springer Berlin Heidelberg.
dc.description.firstpage43
dc.description.lastpage52
dc.description.volume5249 LNAI
dc.identifier.citationTHOMAZ, C. E.; GIRALDI, G. A. Discriminant eigenfaces: A new ranking method for principal components analysis. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), v. 5249 LNAI, p. 43-52, Oct, 2008.
dc.identifier.doi10.1007/978-3-540-88190-2_10
dc.identifier.issn1611-3349
dc.identifier.urihttps://repositorio.fei.edu.br/handle/FEI/4315
dc.relation.ispartofLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
dc.rightsAcesso Restrito
dc.subject.otherlanguageEigenfaces
dc.subject.otherlanguagePrincipal components analysis
dc.subject.otherlanguageSeparating hyperplanes
dc.subject.otherlanguageSmall sample size problems
dc.titleDiscriminant eigenfaces: A new ranking method for principal components analysis
dc.typeArtigo de evento
fei.scopus.citations1
fei.scopus.eid2-s2.0-57049123845
fei.scopus.subjectCovariance structures
fei.scopus.subjectData representations
fei.scopus.subjectEigenfaces
fei.scopus.subjectFirst principal components
fei.scopus.subjectPrincipal components analysis
fei.scopus.subjectRobust reconstruction
fei.scopus.subjectSeparating hyperplane
fei.scopus.subjectSmall sample size problems
fei.scopus.updated2024-07-01
fei.scopus.urlhttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=57049123845&origin=inward
Arquivos
Coleções
Artigos