Recognition of Human Iris Using Accurate Iris Map

Abstract

Personal identification based on Biometrics technology is a trend in future. Iris Recognition is regarded as a high accuracy verification technology when compared to traditional approaches. In real-time Iris Recognition, Iris Localization is very important step for Iris Recognition. So, the Iris Masks plays an important role in iris recognition. This Iris Mask indicates which part is occluded or contaminated by noisy image artifacts such as eyelashes, eyelids, eyeglasses frames, and specular reflections. If the Iris Mask is inaccurate, it will decrease its performance in Iris Recognition system. So, we use learning- based algorithms to estimate accurate Iris Masks from Iris Images, which propose to use Figueiredo and Jain's Gaussian Mixture Models (FJ-GMMs) to model the underlying probabilistic distributions of both valid and invalid regions on iris images. We also explored possible features and found that Gabor Filter Bank (GFB) provides the most discriminative information for our goal. Finally, we applied Simulated Annealing (SA) technique to optimize the parameters of GFB in order to achieve the best recognition rate. Index Terms: Gaussian mixture models, iris mask, iris recognition, iris occlusion estimation, simulated annealing.