Hybrid Fusion based Multimodal Biometric System for Accurate Personal Identification
In the recent years, biometric authentication has become popular in modern society. The recognition accuracy of unimodal biometric systems has to contend with a variety of issues such as background noise, noisy data, non-universality, spoof attacks, intra-class variations, inter-class similarities or distinctiveness, interoperability problems. To overcome the limitation of a single biometrics, information from multiple biometrics can be integrated to achieve more reliable and robust performance. In existing system, score level fusion method is introduced to achieve better identification result using Left and Right Palmprint Images. However, various normalization methods of the matching scores cause different decision boundaries. Also, a too small training set of scores might easily overfits the data, especially in methods with flexible boundaries. To solve this problem, the proposed system introduced a hybrid fusion approach which integrate both score level and decision level fusion. In this proposed system, left and right palmprint of the same subject is correlated and crossing matching score of the left and right palmprint is computed for improving the efficiency of identity identification. Then ROC is derived from the component matching scores and the score-level fused matching scores. Finally combined both score level and decision level results to achieve hybrid fusion. The experimental results show that the proposed system achieves better performance compared with existing system in terms of detection rate and false acceptance rate.