Video Based Non-Cooperative Iris Segmentation

Iris segmentation is one of the most important steps in an iris recognition system. Its accuracy can directly affect the recognition accuracy. For non-cooperative users, the obtained images often do not have good quality. Under such conditions, the iris may be deformed, out-of-focus, or motion blurred. Sometimes, images do not have a valid iris. It is very challenging to segment the iris efficiently and accurately under a non-cooperative scenario. In this paper, we proposed a novel segmentation method that uses a coarse to fine approach to extract the iris region. The preliminary result shows the proposed method is efficient and accurate.