A Robust Iris Segmentation Algorithm Using Active Contours Without Edges and Improved Circular Hough Transform

Iris segmentation plays the most important role in iris biometric system and it determines the subsequent recognizing result. So far, there are still many challenges in this research filed. This paper proposes a robust iris segmentation algorithm using active contours without edges and improved circular Hough transform. Firstly, we adopt a simple linear interpolation model to remove the specular reflections. Secondly, we combine HOG features and Adaboost cascade detector to extract the region of interest from the original iris image. Thirdly, the active contours without edges model and the improved circular Hough transform model are used for the pupillary and limbic boundaries localization, respectively. Lastly, two iris databases CASIA-IrisV1 and CASIA-IrisV4-Lamp were adopted to prove the efficacy of the proposed method. The experimental results show that the performance of proposed method is effective and robust.