Abstract
Biometric fusion is an essential procedure in any multi-modal biometric person recognition system, and it can be performed at sensor, feature, score and decision levels. This paper proposes to use a simulated annealing (SA) algorithm for the integration of multi-modal biometric data. The proposed method is applied to a Audio-Visual (AV) person recognition database that includes both acoustic and visual information. Superior performance of the proposed method is demonstrated through experimentation on 40 subjects in an AV database.