An intelligent decision combiner applied to noncooperative iris recognition

In despite of successful implementation of iris recognition systems, noncooperative recognition is still remained as an unsolved problem. Unexpected behavior of the subjects and uncontrolled lighting conditions as the main aspects of noncooperative iris recognition result in blurred and noisy captured images. This issue can degrade the performance of iris recognition system. In this paper, to address the aforementioned challenges, an intelligent decision combiner is proposed in which prior to perform decision fusion; an automatic image quality inspection is carried out. The goal is to determine whether captured decisions based on visible light (VL) and near infrared (NIR) images have enough reliability to incorporate into final decision making. Experimental results on the UTIRIS confirm the superior performance of the proposed combiner in comparison with other common nontrainable decision combiners whereas in all cases, the effectiveness of fusion approach makes it a reliable solution to noncooperative subjectspsila behavior and uncontrolled lighting conditions.