Skip to main content
Conference ProceedingsOPEN ACCESS

Optimal decision fusion for a face verification system

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (2007) 4642 LNCS 958-967
DOI: 10.1007/978-3-540-74549-5_100
7Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

Fusion is a popular practice to increase the reliability of the biometric verification. In this paper, optimal fusion at decision level by AND rule and OR rule is investigated. Both a theoretical analysis and the experimental results are given. Comparisons are presented between fusion at decision level and fusion at matching score level. For our face verification system, decision fusion proves to be a simple, practical, and effective approach, which significantly improves the performance of the original classifier. © Springer-Verlag Berlin Heidelberg 2007.

Cite

CITATION STYLE

APA

Tao, Q., & Veldhuis, R. (2007). Optimal decision fusion for a face verification system. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4642 LNCS, pp. 958–967). Springer Verlag. https://doi.org/10.1007/978-3-540-74549-5_100

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free