An efficient method to improve iris recognition performance based on multi-level fusion

摘要

To improve the reliability and accuracy of iris-based identification system, this paper proposed an efficient method based on multi-level fusion. Multi-level fusion contains two levels: multialgorithmic fusion and multiinstance fusion. The multialgorithmic fusion integrates the improved phase algorithm and the DCT-based algorithm, and the multiinstance fusion combines information from left iris and right iris of an individual. Both multialgorithmic fusion and multiinstance fusion are carried out at the matching score level. The experimental results on the noisy iris database UBIRIS demonstrate that the proposed fusion schemes can perform better than the simple recognition systems.