PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION

PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION

This paper deals with an original method of structure parametric optimization for multimodal decision-level fusion scheme which combines the results of the partial solution for the classification task obtained from assembly of the monomodal classifiers. As a result, a multimodal fusion classifier wh...

Full description

Bibliographic Details
Main Author: A. V. Timofeev
Format: Article
Language:English
Published: Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University) 2014-05-01
Series:Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki
Subjects:
consolidating classification decision
minimum of classification error
exponential losses function
Online Access:http://ntv.ifmo.ru/file/article/9652.pdf
id doaj-cdda8151fcd24899aaaca8588bde0e3d
record_format Article
spelling doaj-cdda8151fcd24899aaaca8588bde0e3d2020-11-24T21:23:37ZengSaint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University)Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki2226-14942500-03732014-05-0114396102PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATIONA. V. TimofeevThis paper deals with an original method of structure parametric optimization for multimodal decision-level fusion scheme which combines the results of the partial solution for the classification task obtained from assembly of the monomodal classifiers. As a result, a multimodal fusion classifier which has the minimum value of the total error rate has been obtained. Properties of the proposed approach are proved rigorously. Suggested method has an urgent practical application in the automatic multimodal biometric person’s identification systems and in the systems for remote monitoring of extended objects. The proposed solution is easy for practical implementation into real operating systems. The paper presents a simulation study of the effectiveness of this optimized multimodal fusion classifier carried out on special bimodal biometrical database. Simulation results showed high practical effectiveness of the suggested method.http://ntv.ifmo.ru/file/article/9652.pdfconsolidating classification decisionminimum of classification errorexponential losses function
collection DOAJ
language English
format Article
sources DOAJ
author A. V. Timofeev
spellingShingle A. V. Timofeev
PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION
Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki
consolidating classification decision
minimum of classification error
exponential losses function
author_facet A. V. Timofeev
author_sort A. V. Timofeev
title PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION
title_short PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION
title_full PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION
title_fullStr PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION
title_full_unstemmed PARAMETRIC OPTIMIZATION OF THE MULTIMODAL DECISION-LEVEL FUSION SCHEME IN AUTOMATIC BIOMETRIC PERSON’S IDENTIFICATION
title_sort parametric optimization of the multimodal decision-level fusion scheme in automatic biometric person’s identification
publisher Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University)
series Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki
issn 2226-1494
2500-0373
publishDate 2014-05-01
description This paper deals with an original method of structure parametric optimization for multimodal decision-level fusion scheme which combines the results of the partial solution for the classification task obtained from assembly of the monomodal classifiers. As a result, a multimodal fusion classifier which has the minimum value of the total error rate has been obtained. Properties of the proposed approach are proved rigorously. Suggested method has an urgent practical application in the automatic multimodal biometric person’s identification systems and in the systems for remote monitoring of extended objects. The proposed solution is easy for practical implementation into real operating systems. The paper presents a simulation study of the effectiveness of this optimized multimodal fusion classifier carried out on special bimodal biometrical database. Simulation results showed high practical effectiveness of the suggested method.
topic consolidating classification decision
minimum of classification error
exponential losses function
url http://ntv.ifmo.ru/file/article/9652.pdf
work_keys_str_mv AT avtimofeev parametricoptimizationofthemultimodaldecisionlevelfusionschemeinautomaticbiometricpersonsidentification
_version_ 1725991957278228480

Similar Items