Wavelet packet correlation methods in biometrics

Pablo Hennings, Jason Thornton, Jelena Kovacevic, B. V.K.Vijaya Kumar

Research output: Contribution to journalArticle

Abstract

We introduce wavelet packet correlation filter classifiers. Correlation filters are traditionally designed in the image domain by minimization of some criterion function of the image training set. Instead, we perform classification in wavelet spaces that have training set representations that provide better solutions to the optimization problem in the filter design. We propose a pruning algorithm to find these wavelet spaces by using a correlation energy cost function, and we describe a match score fusion algorithm for applying the filters trained across the packet tree. The proposed classification algorithm is suitable for any object-recognition task. We present results by implementing a biometric recognition system that uses the NIST 24 fingerprint database, and show that applying correlation filters in the wavelet domain results in considerable improvement of the standard correlation filter algorithm

Original languageEnglish (US)
Pages (from-to)637-646
Number of pages10
JournalApplied Optics
Volume44
Issue number5
DOIs
StatePublished - Feb 10 2005

Fingerprint

biometrics
Correlation methods
Biometrics
filters
Object recognition
education
Cost functions
optimization
Classifiers
Fusion reactions
classifiers
fusion
costs

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics

Cite this

Hennings, P., Thornton, J., Kovacevic, J., & Kumar, B. V. K. V. (2005). Wavelet packet correlation methods in biometrics. Applied Optics, 44(5), 637-646. https://doi.org/10.1364/AO.44.000637

Wavelet packet correlation methods in biometrics. / Hennings, Pablo; Thornton, Jason; Kovacevic, Jelena; Kumar, B. V.K.Vijaya.

In: Applied Optics, Vol. 44, No. 5, 10.02.2005, p. 637-646.

Research output: Contribution to journalArticle

Hennings, P, Thornton, J, Kovacevic, J & Kumar, BVKV 2005, 'Wavelet packet correlation methods in biometrics', Applied Optics, vol. 44, no. 5, pp. 637-646. https://doi.org/10.1364/AO.44.000637
Hennings, Pablo ; Thornton, Jason ; Kovacevic, Jelena ; Kumar, B. V.K.Vijaya. / Wavelet packet correlation methods in biometrics. In: Applied Optics. 2005 ; Vol. 44, No. 5. pp. 637-646.
@article{5c90a7601b2b4637add09e21e9b05352,
title = "Wavelet packet correlation methods in biometrics",
abstract = "We introduce wavelet packet correlation filter classifiers. Correlation filters are traditionally designed in the image domain by minimization of some criterion function of the image training set. Instead, we perform classification in wavelet spaces that have training set representations that provide better solutions to the optimization problem in the filter design. We propose a pruning algorithm to find these wavelet spaces by using a correlation energy cost function, and we describe a match score fusion algorithm for applying the filters trained across the packet tree. The proposed classification algorithm is suitable for any object-recognition task. We present results by implementing a biometric recognition system that uses the NIST 24 fingerprint database, and show that applying correlation filters in the wavelet domain results in considerable improvement of the standard correlation filter algorithm",
author = "Pablo Hennings and Jason Thornton and Jelena Kovacevic and Kumar, {B. V.K.Vijaya}",
year = "2005",
month = "2",
day = "10",
doi = "10.1364/AO.44.000637",
language = "English (US)",
volume = "44",
pages = "637--646",
journal = "Applied Optics",
issn = "0003-6935",
publisher = "The Optical Society",
number = "5",

}

TY - JOUR

T1 - Wavelet packet correlation methods in biometrics

AU - Hennings, Pablo

AU - Thornton, Jason

AU - Kovacevic, Jelena

AU - Kumar, B. V.K.Vijaya

PY - 2005/2/10

Y1 - 2005/2/10

N2 - We introduce wavelet packet correlation filter classifiers. Correlation filters are traditionally designed in the image domain by minimization of some criterion function of the image training set. Instead, we perform classification in wavelet spaces that have training set representations that provide better solutions to the optimization problem in the filter design. We propose a pruning algorithm to find these wavelet spaces by using a correlation energy cost function, and we describe a match score fusion algorithm for applying the filters trained across the packet tree. The proposed classification algorithm is suitable for any object-recognition task. We present results by implementing a biometric recognition system that uses the NIST 24 fingerprint database, and show that applying correlation filters in the wavelet domain results in considerable improvement of the standard correlation filter algorithm

AB - We introduce wavelet packet correlation filter classifiers. Correlation filters are traditionally designed in the image domain by minimization of some criterion function of the image training set. Instead, we perform classification in wavelet spaces that have training set representations that provide better solutions to the optimization problem in the filter design. We propose a pruning algorithm to find these wavelet spaces by using a correlation energy cost function, and we describe a match score fusion algorithm for applying the filters trained across the packet tree. The proposed classification algorithm is suitable for any object-recognition task. We present results by implementing a biometric recognition system that uses the NIST 24 fingerprint database, and show that applying correlation filters in the wavelet domain results in considerable improvement of the standard correlation filter algorithm

UR - http://www.scopus.com/inward/record.url?scp=14844325734&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=14844325734&partnerID=8YFLogxK

U2 - 10.1364/AO.44.000637

DO - 10.1364/AO.44.000637

M3 - Article

C2 - 15751845

AN - SCOPUS:14844325734

VL - 44

SP - 637

EP - 646

JO - Applied Optics

JF - Applied Optics

SN - 0003-6935

IS - 5

ER -