<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<identifier>6qtX3pFwXQZeBBx/GEG3a</identifier>
		<repository>sid.inpe.br/banon/2005/07.05.16.47</repository>
		<metadatarepository>sid.inpe.br/banon/2005/07.05.16.47.10</metadatarepository>
		<site>sibgrapi.sid.inpe.br 802</site>
		<citationkey>ThomazGill:2005:ApFaRe</citationkey>
		<author>Thomaz, Carlos Eduardo,</author>
		<author>Gillies, Duncan Fyfe,</author>
		<affiliation>Centro Universitario da FEI, Sao Paulo, Brazil,</affiliation>
		<affiliation>Imperial College, London, UK,</affiliation>
		<title>A maximum uncertainty LDA-based approach for limited sample size problems - with application to face recognition</title>
		<conferencename>Brazilian Symposium on Computer Graphics and Image Processing, 18 (SIBGRAPI)</conferencename>
		<year>2005</year>
		<editor>Rodrigues, Maria Andr?ia Formico,</editor>
		<editor>Frery, Alejandro C?sar,</editor>
		<booktitle>Proceedings</booktitle>
		<date>9-12 Oct. 2005</date>
		<publisheraddress>Los Alamitos</publisheraddress>
		<publisher>IEEE Computer Society</publisher>
		<conferencelocation>Natal</conferencelocation>
		<keywords>LDA, maximum uncertainty LDA, limited sample size, face recognition.</keywords>
		<abstract>A critical issue of applying Linear Discriminant Analysis (LDA) is both the singularity and instability of the within-class scatter matrix.  In practice, particularly in image recognition applications such as face recognition, there are often a large number of pixels or pre-processed features available, but the total number of training patterns is limited and commonly less than the dimension of the feature space.  In this paper, a maximum uncertainty LDA-based method is proposed. It is based on a straightforward stabilisation approach for the within-class scatter matrix.  In order to evaluate its effectiveness, experiments on face recognition using the well-known ORL and FERET face databases were carried out and compared with other LDA-based methods.  The results indicate that our method improves the LDA classification performance when the within-class scatter matrix is not only singular but also poorly estimated, with or without a Principal Component Analysis intermediate step and using less linear discriminant features.</abstract>
		<language>en</language>
		<tertiarytype>Full Paper</tertiarytype>
		<format>On-line</format>
		<size>107 KiB</size>
		<numberoffiles>1</numberoffiles>
		<targetfile>thomazc_mlda.pdf</targetfile>
		<lastupdate>2005:07.05.03.00.00 sid.inpe.br/banon/2001/03.30.15.38 administrator</lastupdate>
		<metadatalastupdate>2020:02.19.03.19.08 sid.inpe.br/banon/2001/03.30.15.38 administrator {D 2005}</metadatalastupdate>
		<e-mailaddress>cet@fei.edu.br</e-mailaddress>
		<usergroup>cet administrator</usergroup>
		<visibility>shown</visibility>
		<transferableflag>1</transferableflag>
		<hostcollection>sid.inpe.br/banon/2001/03.30.15.38</hostcollection>
		<contenttype>External Contribution</contenttype>
		<lasthostcollection>sid.inpe.br/banon/2001/03.30.15.38</lasthostcollection>
		<url>http://sibgrapi.sid.inpe.br/rep-/sid.inpe.br/banon/2005/07.05.16.47</url>
	</metadata>
</metadatalist>