Two-fold regularization for kernel Fisher discriminant analysis in face recognition

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Due to the inherent nature of kernel implementation, the kernel Fisher discriminant suffers from the small sample size problem. In this paper, we introduce a novel variant of the kernel Fisher discriminant formulation to circumvent this problem. By adopting a two-fold regularization scheme on the scatter matrices, we show both effectiveness and reliability of the proposed method particularly regarding the small sample size and the lack of dimensionality issues.

Original languageEnglish
Pages (from-to)540-545
Number of pages6
Journalieice electronics express
Volume6
Issue number9
DOIs
Publication statusPublished - 2009 May 10

Fingerprint

Discriminant analysis
Face recognition
formulations
matrices

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Electrical and Electronic Engineering

Cite this

@article{f481478d04e5490dbe625b2e6ef28438,
title = "Two-fold regularization for kernel Fisher discriminant analysis in face recognition",
abstract = "Due to the inherent nature of kernel implementation, the kernel Fisher discriminant suffers from the small sample size problem. In this paper, we introduce a novel variant of the kernel Fisher discriminant formulation to circumvent this problem. By adopting a two-fold regularization scheme on the scatter matrices, we show both effectiveness and reliability of the proposed method particularly regarding the small sample size and the lack of dimensionality issues.",
author = "Kim, {Sang Ki} and Toh, {Kar Ann} and Lee, {Sang Youn}",
year = "2009",
month = "5",
day = "10",
doi = "10.1587/elex.6.540",
language = "English",
volume = "6",
pages = "540--545",
journal = "IEICE Electronics Express",
issn = "1349-2543",
publisher = "The Institute of Electronics, Information and Communication Engineers (IEICE)",
number = "9",

}

Two-fold regularization for kernel Fisher discriminant analysis in face recognition. / Kim, Sang Ki; Toh, Kar Ann; Lee, Sang Youn.

In: ieice electronics express, Vol. 6, No. 9, 10.05.2009, p. 540-545.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Two-fold regularization for kernel Fisher discriminant analysis in face recognition

AU - Kim, Sang Ki

AU - Toh, Kar Ann

AU - Lee, Sang Youn

PY - 2009/5/10

Y1 - 2009/5/10

N2 - Due to the inherent nature of kernel implementation, the kernel Fisher discriminant suffers from the small sample size problem. In this paper, we introduce a novel variant of the kernel Fisher discriminant formulation to circumvent this problem. By adopting a two-fold regularization scheme on the scatter matrices, we show both effectiveness and reliability of the proposed method particularly regarding the small sample size and the lack of dimensionality issues.

AB - Due to the inherent nature of kernel implementation, the kernel Fisher discriminant suffers from the small sample size problem. In this paper, we introduce a novel variant of the kernel Fisher discriminant formulation to circumvent this problem. By adopting a two-fold regularization scheme on the scatter matrices, we show both effectiveness and reliability of the proposed method particularly regarding the small sample size and the lack of dimensionality issues.

UR - http://www.scopus.com/inward/record.url?scp=67249143949&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67249143949&partnerID=8YFLogxK

U2 - 10.1587/elex.6.540

DO - 10.1587/elex.6.540

M3 - Article

VL - 6

SP - 540

EP - 545

JO - IEICE Electronics Express

JF - IEICE Electronics Express

SN - 1349-2543

IS - 9

ER -