Iris Recognition Based on Local Binary Descriptors

The use of biological properties for individual identification, called biometric systems, on mobile devices is the easier and safer approach to deal with user personal information. Several works have been sought to develop robust solutions for different biometric modalities, such as, face, fingerprint, palmprint, voice, and iris recognition. In this work, we evaluate different local binary descriptors, which have fast processing and still efficient memory usage for iris recognition task. At the best of our knowledge, this is the first system that uses local binary descriptors for this task. In our experiments, we compare three wellknown local binary descriptors - BRIEF, ORB and BRISK - in a criterious experimental analysis that considers effectiveness, time-consuming, and memory usage metrics.