Analysis of Iris Images Segmentation Methods Based on Level Set

Iris recognition plays an important role in personal identification. In this study, we utilized CREASEG experimental platform to analyze the performance of some state-of-the-art image segmentation algorithms based on level set. Performance evaluation criteria include segmentation accuracy and computation time of pupil and iris localization. Four iris images were taken as experimental samples. The experimental results on those image samples demonstrate that Chan-Vese model achieve the best performance among all six algorithms. Furthermore, experimental results also show that energy functions play an important role, which should not make evolution curve to terminate at local minima or pass through the boundary. This study can provide certain referential significance in how to select image segmentation algorithm based on level set.