Random projection with robust linear discriminant analysis model in face recognition

Pang Ying Han, Andrew Teoh Beng Jin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents a face recognition technique with two techniques: random projection (RP) and Robust linear Discriminant analysis Model (RDM). RDM is an enhanced version of Fisher's Linear Discriminant with energy-adaptive regularization criteria. It is able to yield better discrimination performance. Same as Fisher's Linear Discriminant, it also faces the singularity problem of within-class scatter. Thus, a dimensionality reduction technique, such as Principal Component Analsys (PCA), is needed to deal with this problem. In this paper, RP is used as an alternative to PCA in RDM in the application of face recognition. Unlike PCA, RP is training data independent and the random subspace computation is relatively simple. The experimental results illustrate that the proposed algorithm is able to attain better recognition performance (error rate is approximately 5% lower) compared to Fisherfaces.

Original languageEnglish
Title of host publicationComputer Graphics, Imaging and Visualisation
Subtitle of host publicationNew Advances, CGIV 2007
Pages11-15
Number of pages5
DOIs
Publication statusPublished - 2007 Dec 1
EventComputer Graphics, Imaging and Visualisation: New Advances, CGIV 2007 - Bangkok, Thailand
Duration: 2007 Aug 132007 Aug 16

Publication series

NameComputer Graphics, Imaging and Visualisation: New Advances, CGIV 2007

Other

OtherComputer Graphics, Imaging and Visualisation: New Advances, CGIV 2007
CountryThailand
CityBangkok
Period07/8/1307/8/16

Fingerprint

Discriminant analysis
Face recognition

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Cite this

Han, P. Y., & Jin, A. T. B. (2007). Random projection with robust linear discriminant analysis model in face recognition. In Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007 (pp. 11-15). [4293642] (Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007). https://doi.org/10.1109/CGIV.2007.70
Han, Pang Ying ; Jin, Andrew Teoh Beng. / Random projection with robust linear discriminant analysis model in face recognition. Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007. 2007. pp. 11-15 (Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007).
@inproceedings{2251eedf4e7a4bf3999ca38246e7d237,
title = "Random projection with robust linear discriminant analysis model in face recognition",
abstract = "This paper presents a face recognition technique with two techniques: random projection (RP) and Robust linear Discriminant analysis Model (RDM). RDM is an enhanced version of Fisher's Linear Discriminant with energy-adaptive regularization criteria. It is able to yield better discrimination performance. Same as Fisher's Linear Discriminant, it also faces the singularity problem of within-class scatter. Thus, a dimensionality reduction technique, such as Principal Component Analsys (PCA), is needed to deal with this problem. In this paper, RP is used as an alternative to PCA in RDM in the application of face recognition. Unlike PCA, RP is training data independent and the random subspace computation is relatively simple. The experimental results illustrate that the proposed algorithm is able to attain better recognition performance (error rate is approximately 5{\%} lower) compared to Fisherfaces.",
author = "Han, {Pang Ying} and Jin, {Andrew Teoh Beng}",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/CGIV.2007.70",
language = "English",
isbn = "0769529283",
series = "Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007",
pages = "11--15",
booktitle = "Computer Graphics, Imaging and Visualisation",

}

Han, PY & Jin, ATB 2007, Random projection with robust linear discriminant analysis model in face recognition. in Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007., 4293642, Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007, pp. 11-15, Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007, Bangkok, Thailand, 07/8/13. https://doi.org/10.1109/CGIV.2007.70

Random projection with robust linear discriminant analysis model in face recognition. / Han, Pang Ying; Jin, Andrew Teoh Beng.

Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007. 2007. p. 11-15 4293642 (Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Random projection with robust linear discriminant analysis model in face recognition

AU - Han, Pang Ying

AU - Jin, Andrew Teoh Beng

PY - 2007/12/1

Y1 - 2007/12/1

N2 - This paper presents a face recognition technique with two techniques: random projection (RP) and Robust linear Discriminant analysis Model (RDM). RDM is an enhanced version of Fisher's Linear Discriminant with energy-adaptive regularization criteria. It is able to yield better discrimination performance. Same as Fisher's Linear Discriminant, it also faces the singularity problem of within-class scatter. Thus, a dimensionality reduction technique, such as Principal Component Analsys (PCA), is needed to deal with this problem. In this paper, RP is used as an alternative to PCA in RDM in the application of face recognition. Unlike PCA, RP is training data independent and the random subspace computation is relatively simple. The experimental results illustrate that the proposed algorithm is able to attain better recognition performance (error rate is approximately 5% lower) compared to Fisherfaces.

AB - This paper presents a face recognition technique with two techniques: random projection (RP) and Robust linear Discriminant analysis Model (RDM). RDM is an enhanced version of Fisher's Linear Discriminant with energy-adaptive regularization criteria. It is able to yield better discrimination performance. Same as Fisher's Linear Discriminant, it also faces the singularity problem of within-class scatter. Thus, a dimensionality reduction technique, such as Principal Component Analsys (PCA), is needed to deal with this problem. In this paper, RP is used as an alternative to PCA in RDM in the application of face recognition. Unlike PCA, RP is training data independent and the random subspace computation is relatively simple. The experimental results illustrate that the proposed algorithm is able to attain better recognition performance (error rate is approximately 5% lower) compared to Fisherfaces.

UR - http://www.scopus.com/inward/record.url?scp=46449102697&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46449102697&partnerID=8YFLogxK

U2 - 10.1109/CGIV.2007.70

DO - 10.1109/CGIV.2007.70

M3 - Conference contribution

AN - SCOPUS:46449102697

SN - 0769529283

SN - 9780769529288

T3 - Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007

SP - 11

EP - 15

BT - Computer Graphics, Imaging and Visualisation

ER -

Han PY, Jin ATB. Random projection with robust linear discriminant analysis model in face recognition. In Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007. 2007. p. 11-15. 4293642. (Computer Graphics, Imaging and Visualisation: New Advances, CGIV 2007). https://doi.org/10.1109/CGIV.2007.70