Evolutionary Fusion of a Multi-Classifier System for Efficient Face Recognition

In this paper an evolutionary classifier fusion method inspired by biological evolution is presented to optimize the performance of a face recognition system. Initially, different illumination environments are modeled as multiple contexts using unsupervised learning and then the optimized classifier...

Full description

Saved in:
Bibliographic Details
Published inInternational journal of control, automation, and systems Vol. 7; no. 1; pp. 33 - 40
Main Authors Yu, Zhan, Nam, Mi-Young, Sedai, Suman, Rhee, Phill-Kyu
Format Journal Article
LanguageKorean
Published 2009
Subjects
evolutionary fusion
multiple context
Classifier fusion
classifier selection
Online AccessGet full text

Cover

More Information
Summary:In this paper an evolutionary classifier fusion method inspired by biological evolution is presented to optimize the performance of a face recognition system. Initially, different illumination environments are modeled as multiple contexts using unsupervised learning and then the optimized classifier ensemble is searched for each context using a Genetic Algorithm (GA). For each context, multiple optimized classifiers are searched; each of which are referred to as a context based classifier. An evolutionary framework comprised of a combination of these classifiers is then applied to optimize face recognition as a whole. Evolutionary classifier fusion is compared with the simple adaptive system. Experiments are carried out using the Inha database and FERET database. Experimental results show that the proposed evolutionary classifier fusion method gives superior performance over other methods without using evolutionary fusion.
Bibliography:KISTI1.1003/JNL.JAKO200907653005643
ISSN:1598-6446
2005-4092