Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
Multispace KL for pattern representation and classification
Autore:
Cappelli, R; Maio, D; Maltoni, D;
Indirizzi:
Univ Bologna, Cdl Sci Informaz, I-47023 Cesena, FO, Italy Univ Bologna Cesena FO Italy I-47023 Informaz, I-47023 Cesena, FO, Italy Univ Bologna, DEIS, CSITE CNR, I-40136 Bologna, Italy Univ Bologna Bologna Italy I-40136 IS, CSITE CNR, I-40136 Bologna, Italy
Titolo Testata:
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
fascicolo: 9, volume: 23, anno: 2001,
pagine: 977 - 996
SICI:
0162-8828(200109)23:9<977:MKFPRA>2.0.ZU;2-L
Fonte:
ISI
Lingua:
ENG
Soggetto:
HOUGH TRANSFORM; IMAGE RETRIEVAL; HUMAN FACES; ALGORITHM; VISION;
Keywords:
KL transform; PCA; multispace KL; clustering; piecewise-linear approximation; face representation; face recognition;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
28
Recensione:
Indirizzi per estratti:
Indirizzo: Cappelli, R Univ Bologna, Cdl Sci Informaz, Via Sacchi 3, I-47023 Cesena, FO, Italy Univ Bologna Via Sacchi 3 Cesena FO Italy I-47023 a, FO, Italy
Citazione:
R. Cappelli et al., "Multispace KL for pattern representation and classification", IEEE PATT A, 23(9), 2001, pp. 977-996

Abstract

This work introduces the Multispace KL (MKL) as a new approach to unsupervised dimensionality reduction for pattern representation and classification. The training set is automatically partitioned into disjoint subsets, according to an optimality criterion; each subset then determines a different KL subspace which is specialized in representing a particular group of patterns. The extension of the classical KL operators and the definition of ad hoc distances allow MKL to be effectively used where KL is commonly employed. The limits of the standard KL transform are pointed out, in particular, MKL is shown to outperform KL when the data distribution is far from a multidimensional Gaussian and to better cope with large sets of patterns, which could cause a severe performance drop in KL.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 29/01/20 alle ore 19:06:03