Extraction and fusion of partial face features for cancelable identity verification

Beom Seok Oh, Kar Ann Toh, Kwontaeg Choi, Beng Jin Teoh, Jaihie Kim

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

In this paper, we propose to extract localized random features directly from partial face image matrix for cancelable identity verification. Essentially, the extracted random features consist of compressed horizontal and vertical facial information obtained from a structured projection of the raw face images. For template security reason, the face appearance information is concealed via averaging several templates over different transformations. The match score outputs of these cancelable templates are then fused through a total error rate minimization. Extensive experiments were carried out to evaluate and benchmark the performance of the proposed method based on the AR, FERET, ORL, Sheffield and BERC databases. Our empirical results show encouraging performances in terms of verification accuracy as well as satisfying four cancelable biometric properties.

Original languageEnglish
Pages (from-to)3288-3303
Number of pages16
JournalPattern Recognition
Volume45
Issue number9
DOIs
Publication statusPublished - 2012 Sep 1

Fingerprint

Fusion reactions
Biometrics
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

@article{e0ab745784764e56b9f91529f05e12a0,
title = "Extraction and fusion of partial face features for cancelable identity verification",
abstract = "In this paper, we propose to extract localized random features directly from partial face image matrix for cancelable identity verification. Essentially, the extracted random features consist of compressed horizontal and vertical facial information obtained from a structured projection of the raw face images. For template security reason, the face appearance information is concealed via averaging several templates over different transformations. The match score outputs of these cancelable templates are then fused through a total error rate minimization. Extensive experiments were carried out to evaluate and benchmark the performance of the proposed method based on the AR, FERET, ORL, Sheffield and BERC databases. Our empirical results show encouraging performances in terms of verification accuracy as well as satisfying four cancelable biometric properties.",
author = "Oh, {Beom Seok} and Toh, {Kar Ann} and Kwontaeg Choi and Teoh, {Beng Jin} and Jaihie Kim",
year = "2012",
month = "9",
day = "1",
doi = "10.1016/j.patcog.2012.02.027",
language = "English",
volume = "45",
pages = "3288--3303",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "9",

}

Extraction and fusion of partial face features for cancelable identity verification. / Oh, Beom Seok; Toh, Kar Ann; Choi, Kwontaeg; Teoh, Beng Jin; Kim, Jaihie.

In: Pattern Recognition, Vol. 45, No. 9, 01.09.2012, p. 3288-3303.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Extraction and fusion of partial face features for cancelable identity verification

AU - Oh, Beom Seok

AU - Toh, Kar Ann

AU - Choi, Kwontaeg

AU - Teoh, Beng Jin

AU - Kim, Jaihie

PY - 2012/9/1

Y1 - 2012/9/1

N2 - In this paper, we propose to extract localized random features directly from partial face image matrix for cancelable identity verification. Essentially, the extracted random features consist of compressed horizontal and vertical facial information obtained from a structured projection of the raw face images. For template security reason, the face appearance information is concealed via averaging several templates over different transformations. The match score outputs of these cancelable templates are then fused through a total error rate minimization. Extensive experiments were carried out to evaluate and benchmark the performance of the proposed method based on the AR, FERET, ORL, Sheffield and BERC databases. Our empirical results show encouraging performances in terms of verification accuracy as well as satisfying four cancelable biometric properties.

AB - In this paper, we propose to extract localized random features directly from partial face image matrix for cancelable identity verification. Essentially, the extracted random features consist of compressed horizontal and vertical facial information obtained from a structured projection of the raw face images. For template security reason, the face appearance information is concealed via averaging several templates over different transformations. The match score outputs of these cancelable templates are then fused through a total error rate minimization. Extensive experiments were carried out to evaluate and benchmark the performance of the proposed method based on the AR, FERET, ORL, Sheffield and BERC databases. Our empirical results show encouraging performances in terms of verification accuracy as well as satisfying four cancelable biometric properties.

UR - http://www.scopus.com/inward/record.url?scp=84862817362&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862817362&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2012.02.027

DO - 10.1016/j.patcog.2012.02.027

M3 - Article

AN - SCOPUS:84862817362

VL - 45

SP - 3288

EP - 3303

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 9

ER -