Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system

A bimodal biometric verification system based on facial and vocal biometric modules is described in this paper. The system under consideration is built in parallel where each matching score reported by two classifiers are fused by using theoretic evidence k-NN (tekNN) based on Dempster-Safer (D-S) t...

Full description

Bibliographic Details
Main Authors: Jin, , ATB, Hussain,, A, Samad,, SA
Format: Article
Published: 2003
Subjects:
Online Access:http://shdl.mmu.edu.my/2600/
_version_ 1848790099624460288
author Jin, , ATB
Hussain,, A
Samad,, SA
author_facet Jin, , ATB
Hussain,, A
Samad,, SA
author_sort Jin, , ATB
building MMU Institutional Repository
collection Online Access
description A bimodal biometric verification system based on facial and vocal biometric modules is described in this paper. The system under consideration is built in parallel where each matching score reported by two classifiers are fused by using theoretic evidence k-NN (tekNN) based on Dempster-Safer (D-S) theory. In this technique, each nearest neighbour of a pattern to be classified is regarded as an item of evidence supporting certain hypotheses concerning the pattern class membership. Unlike statistical based fusion approaches, tekNN based on D-S theory is able to represent uncertainties and lack of knowledge. Therefore, the usage of tekNN leads to a ternary decision scheme, {accept, reject, inconclusive} which provides a more secure protection. From experimental results, the speech and facial biometric modules perform equally well, giving 93.5% and 94.0% verification rates, respectively. A 99.86% recognition rate is obtained when the two modules are fused. In addition, an 'unbalanced' case is been created to investigate the robustness of technique.
first_indexed 2025-11-14T18:07:14Z
format Article
id mmu-2600
institution Multimedia University
institution_category Local University
last_indexed 2025-11-14T18:07:14Z
publishDate 2003
recordtype eprints
repository_type Digital Repository
spelling mmu-26002011-08-24T00:08:26Z http://shdl.mmu.edu.my/2600/ Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system Jin, , ATB Hussain,, A Samad,, SA QA75.5-76.95 Electronic computers. Computer science A bimodal biometric verification system based on facial and vocal biometric modules is described in this paper. The system under consideration is built in parallel where each matching score reported by two classifiers are fused by using theoretic evidence k-NN (tekNN) based on Dempster-Safer (D-S) theory. In this technique, each nearest neighbour of a pattern to be classified is regarded as an item of evidence supporting certain hypotheses concerning the pattern class membership. Unlike statistical based fusion approaches, tekNN based on D-S theory is able to represent uncertainties and lack of knowledge. Therefore, the usage of tekNN leads to a ternary decision scheme, {accept, reject, inconclusive} which provides a more secure protection. From experimental results, the speech and facial biometric modules perform equally well, giving 93.5% and 94.0% verification rates, respectively. A 99.86% recognition rate is obtained when the two modules are fused. In addition, an 'unbalanced' case is been created to investigate the robustness of technique. 2003 Article NonPeerReviewed Jin, , ATB and Hussain,, A and Samad,, SA (2003) Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system. AUDIO-AND VIDEO-BASED BIOMETRIC PERSON AUTHENTICATION, PROCEEDINGS, 2688 . pp. 778-786. ISSN 0302-9743
spellingShingle QA75.5-76.95 Electronic computers. Computer science
Jin, , ATB
Hussain,, A
Samad,, SA
Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system
title Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system
title_full Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system
title_fullStr Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system
title_full_unstemmed Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system
title_short Theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system
title_sort theoretic evidence k-nearest neighbourhood classifiers in a bimodal biometric verification system
topic QA75.5-76.95 Electronic computers. Computer science
url http://shdl.mmu.edu.my/2600/