A novel method for iris recognition based on zero-crossing detection

Abstract

As a kind of biometrics, iris recognition has been developed quickly in recent years. This paper studies a kind of iris identification algorithms which uses the longest-chord method for pupil localization and Hough transform for outer iris boundary location. Then the zero-crossing detection algorithm is used for iris feature extraction, and improved hamming distance is calculated for feature matching between templates of iris. Then, the influence of various parameters on the performance of the iris recognition program is studied deeply. © 2012 Springer-Verlag Berlin Heidelberg.