Research On Location Algorithm Of Iris Recognition

Biometrics is a new identification technology, it recognizes the media as an inherent feature of the human body, and then accurately identifies individual. Iris recognition has a higher accuracy and defensive, which dues to its advantage of the physiological characteristics. Therefore, iris recognition is considered to be a research hotspot, which has broad application prospects. In this paper, it is mainly research the light spots'removal for the iris image and the location for the iris inner and outer boundaries. The research realized the following three points: de-noising, iris location and the normalizing.In the iris location respect, we propose an improved algorithm for iris location, in order to improve the position of the speed, accuracy, robustness. The main features of this algorithm are: (1) because the inner and outer boundaries of iris are nearly a circle, we conducted a round of fitting operations, based on a circle can be fixed by three points which are not in the same line. (2) Part of the iris image of the pupil section contains the light spots, due to the influence of light collection device. It affects the location of the iris inner boundary, resulted in a condition that the location of iris inner boundary is wrong. We use the regional growing method to remove facula's, as to get the correct location of the iris image.In order to eliminate the impact which the iris texture elastic deformation to the follow-up work, we normalized the circle area of the iris which has been located.Finally, we summarize the results and conclusions of the paper, at the same time, some expectations to future research are predicted.