Minimax Probability Machine multialgorithmic fusion for iris recognition

Abstract

This study presents an iris recognition method based on multialgorithmic fusion. Fusion of multiple algorithms for biometric verification performance improvement has received considerable attention. The proposed method combines two kinds of iris recognition algorithms: one is based on phase information and the other is based on zero-crossing representation. Two algorithms are fused at the matching score level and a novel fusion strategy based on Minimax Probability Machine (MPM) is applied to generate a fused score which is used to make the final decision. The experimental results on CASIA and UBIRIS iris image databases show that the proposed multialgorithmic fusion method can bring obvious performance improvement compared to the individual recognition algorithm. The comparison among all fusion strategies also demonstrates that the fusion strategy based on MPM can achieve better performance than traditional fusion strategies. © 2007 Asian Network for Scientific Information.