Fingerprint classification using one-vs-all support vector machines dynamically ordered with nai{dotless}̈ve Bayes classifiers

Jin Hyuk Hong, Jun Ki Min, Ung Keun Cho, Sung Bae Cho

Research output: Contribution to journalArticle

86 Citations (Scopus)

Abstract

Fingerprint classification reduces the number of possible matches in automated fingerprint identification systems by categorizing fingerprints into predefined classes. Support vector machines (SVMs) are widely used in pattern classification and have produced high accuracy when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification systems, we propose a novel method in which the SVMs are generated with the one-vs-all (OVA) scheme and dynamically ordered with nai{dotless}̈ve Bayes classifiers. This is necessary to break the ties that frequently occur when working with multi-class classification systems that use OVA SVMs. More specifically, it uses representative fingerprint features as the FingerCode, singularities and pseudo ridges to train the OVA SVMs and nai{dotless}̈ve Bayes classifiers. The proposed method has been validated on the NIST-4 database and produced a classification accuracy of 90.8% for five-class classification with the statistical significance. The results show the benefits of integrating different fingerprint features as well as the usefulness of the proposed method in multi-class fingerprint classification.

Original languageEnglish
Pages (from-to)662-671
Number of pages10
JournalPattern Recognition
Volume41
Issue number2
DOIs
Publication statusPublished - 2008 Feb 1

Fingerprint

Support vector machines
Classifiers
Pattern recognition
Identification (control systems)

All Science Journal Classification (ASJC) codes

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

Cite this

@article{ef6c5c68678046d0bfd6a26758cd341d,
title = "Fingerprint classification using one-vs-all support vector machines dynamically ordered with nai{dotless}̈ve Bayes classifiers",
abstract = "Fingerprint classification reduces the number of possible matches in automated fingerprint identification systems by categorizing fingerprints into predefined classes. Support vector machines (SVMs) are widely used in pattern classification and have produced high accuracy when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification systems, we propose a novel method in which the SVMs are generated with the one-vs-all (OVA) scheme and dynamically ordered with nai{dotless}̈ve Bayes classifiers. This is necessary to break the ties that frequently occur when working with multi-class classification systems that use OVA SVMs. More specifically, it uses representative fingerprint features as the FingerCode, singularities and pseudo ridges to train the OVA SVMs and nai{dotless}̈ve Bayes classifiers. The proposed method has been validated on the NIST-4 database and produced a classification accuracy of 90.8{\%} for five-class classification with the statistical significance. The results show the benefits of integrating different fingerprint features as well as the usefulness of the proposed method in multi-class fingerprint classification.",
author = "Hong, {Jin Hyuk} and Min, {Jun Ki} and Cho, {Ung Keun} and Cho, {Sung Bae}",
year = "2008",
month = "2",
day = "1",
doi = "10.1016/j.patcog.2007.07.004",
language = "English",
volume = "41",
pages = "662--671",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "2",

}

Fingerprint classification using one-vs-all support vector machines dynamically ordered with nai{dotless}̈ve Bayes classifiers. / Hong, Jin Hyuk; Min, Jun Ki; Cho, Ung Keun; Cho, Sung Bae.

In: Pattern Recognition, Vol. 41, No. 2, 01.02.2008, p. 662-671.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fingerprint classification using one-vs-all support vector machines dynamically ordered with nai{dotless}̈ve Bayes classifiers

AU - Hong, Jin Hyuk

AU - Min, Jun Ki

AU - Cho, Ung Keun

AU - Cho, Sung Bae

PY - 2008/2/1

Y1 - 2008/2/1

N2 - Fingerprint classification reduces the number of possible matches in automated fingerprint identification systems by categorizing fingerprints into predefined classes. Support vector machines (SVMs) are widely used in pattern classification and have produced high accuracy when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification systems, we propose a novel method in which the SVMs are generated with the one-vs-all (OVA) scheme and dynamically ordered with nai{dotless}̈ve Bayes classifiers. This is necessary to break the ties that frequently occur when working with multi-class classification systems that use OVA SVMs. More specifically, it uses representative fingerprint features as the FingerCode, singularities and pseudo ridges to train the OVA SVMs and nai{dotless}̈ve Bayes classifiers. The proposed method has been validated on the NIST-4 database and produced a classification accuracy of 90.8% for five-class classification with the statistical significance. The results show the benefits of integrating different fingerprint features as well as the usefulness of the proposed method in multi-class fingerprint classification.

AB - Fingerprint classification reduces the number of possible matches in automated fingerprint identification systems by categorizing fingerprints into predefined classes. Support vector machines (SVMs) are widely used in pattern classification and have produced high accuracy when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification systems, we propose a novel method in which the SVMs are generated with the one-vs-all (OVA) scheme and dynamically ordered with nai{dotless}̈ve Bayes classifiers. This is necessary to break the ties that frequently occur when working with multi-class classification systems that use OVA SVMs. More specifically, it uses representative fingerprint features as the FingerCode, singularities and pseudo ridges to train the OVA SVMs and nai{dotless}̈ve Bayes classifiers. The proposed method has been validated on the NIST-4 database and produced a classification accuracy of 90.8% for five-class classification with the statistical significance. The results show the benefits of integrating different fingerprint features as well as the usefulness of the proposed method in multi-class fingerprint classification.

UR - http://www.scopus.com/inward/record.url?scp=34848849013&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34848849013&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2007.07.004

DO - 10.1016/j.patcog.2007.07.004

M3 - Article

AN - SCOPUS:34848849013

VL - 41

SP - 662

EP - 671

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 2

ER -