Structural hidden Markov models for biometrics: fusion of face and fingerprint

Djamel Bouchaffra, Abbes Amira

Research output: Contribution to journalArticle

Abstract

The goal of this paper is threefold: (i) propose a novel face and fingerprint feature modeling using the structural hidden Markov models (SHMMs) paradigm, (ii) explore the use of some feature extraction techniques such as ridgelet transform, discrete wavelet transform with various classifiers for biometric identification, and (iii) determine the best method for classifier combination. The experimental results reported in both fingerprint and face recognition reveal that the SHMMs concept is promising since it has outperformed several state-of-the-arts classifiers when combined with the discrete wavelet transform. Besides, this study has shown that the ridgelet transform without principal components analysis (PCA) dimension reduction fits better with the support vector machines (SVMs) classifier than it does with the SHMMs in the fingerprint recognition task. Finally, these results also reveal a small improvement of the bimodal biometric system over unimodal systems; which suggest that a most effective fusion scheme is necessary. (C) 2007 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.
Original languageEnglish
Pages (from-to)852-867
Number of pages16
JournalPattern Recognition
Volume41
Issue number3
DOIs
Publication statusPublished - Mar 2008
Externally publishedYes

Keywords

  • multimodal biometrics
  • discrete wavelet transform
  • ridgelet transform
  • structural hidden Markov models
  • support vector machines
  • classifier combination

Cite this

Bouchaffra, Djamel ; Amira, Abbes. / Structural hidden Markov models for biometrics : fusion of face and fingerprint. In: Pattern Recognition. 2008 ; Vol. 41, No. 3. pp. 852-867.
@article{e5c8865b334843cfb8f88da3225d9ade,
title = "Structural hidden Markov models for biometrics: fusion of face and fingerprint",
abstract = "The goal of this paper is threefold: (i) propose a novel face and fingerprint feature modeling using the structural hidden Markov models (SHMMs) paradigm, (ii) explore the use of some feature extraction techniques such as ridgelet transform, discrete wavelet transform with various classifiers for biometric identification, and (iii) determine the best method for classifier combination. The experimental results reported in both fingerprint and face recognition reveal that the SHMMs concept is promising since it has outperformed several state-of-the-arts classifiers when combined with the discrete wavelet transform. Besides, this study has shown that the ridgelet transform without principal components analysis (PCA) dimension reduction fits better with the support vector machines (SVMs) classifier than it does with the SHMMs in the fingerprint recognition task. Finally, these results also reveal a small improvement of the bimodal biometric system over unimodal systems; which suggest that a most effective fusion scheme is necessary. (C) 2007 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.",
keywords = "multimodal biometrics, discrete wavelet transform, ridgelet transform, structural hidden Markov models, support vector machines, classifier combination",
author = "Djamel Bouchaffra and Abbes Amira",
year = "2008",
month = "3",
doi = "10.1016/j.patcog.2007.06.033",
language = "English",
volume = "41",
pages = "852--867",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier B.V.",
number = "3",

}

Structural hidden Markov models for biometrics : fusion of face and fingerprint. / Bouchaffra, Djamel; Amira, Abbes.

In: Pattern Recognition, Vol. 41, No. 3, 03.2008, p. 852-867.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Structural hidden Markov models for biometrics

T2 - fusion of face and fingerprint

AU - Bouchaffra, Djamel

AU - Amira, Abbes

PY - 2008/3

Y1 - 2008/3

N2 - The goal of this paper is threefold: (i) propose a novel face and fingerprint feature modeling using the structural hidden Markov models (SHMMs) paradigm, (ii) explore the use of some feature extraction techniques such as ridgelet transform, discrete wavelet transform with various classifiers for biometric identification, and (iii) determine the best method for classifier combination. The experimental results reported in both fingerprint and face recognition reveal that the SHMMs concept is promising since it has outperformed several state-of-the-arts classifiers when combined with the discrete wavelet transform. Besides, this study has shown that the ridgelet transform without principal components analysis (PCA) dimension reduction fits better with the support vector machines (SVMs) classifier than it does with the SHMMs in the fingerprint recognition task. Finally, these results also reveal a small improvement of the bimodal biometric system over unimodal systems; which suggest that a most effective fusion scheme is necessary. (C) 2007 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

AB - The goal of this paper is threefold: (i) propose a novel face and fingerprint feature modeling using the structural hidden Markov models (SHMMs) paradigm, (ii) explore the use of some feature extraction techniques such as ridgelet transform, discrete wavelet transform with various classifiers for biometric identification, and (iii) determine the best method for classifier combination. The experimental results reported in both fingerprint and face recognition reveal that the SHMMs concept is promising since it has outperformed several state-of-the-arts classifiers when combined with the discrete wavelet transform. Besides, this study has shown that the ridgelet transform without principal components analysis (PCA) dimension reduction fits better with the support vector machines (SVMs) classifier than it does with the SHMMs in the fingerprint recognition task. Finally, these results also reveal a small improvement of the bimodal biometric system over unimodal systems; which suggest that a most effective fusion scheme is necessary. (C) 2007 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

KW - multimodal biometrics

KW - discrete wavelet transform

KW - ridgelet transform

KW - structural hidden Markov models

KW - support vector machines

KW - classifier combination

U2 - 10.1016/j.patcog.2007.06.033

DO - 10.1016/j.patcog.2007.06.033

M3 - Article

VL - 41

SP - 852

EP - 867

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 3

ER -