Fast iris boundary localization algorithm supervised by pupil center

The accuracy and the speed of iris boundary localization affect the recognition system performance. Hough transform is a common algorithm of geometry object detection in computer vision, and it can be used for iris boundary localization by voting for parameters of circles in eye images. Because the position and the size of irises in images are not fixed, these reasons result in that the computing burden increases and the iris boundary localization is slow. Due to unideal real-time character of iris boundary localization, a fast iris boundary localization algorithm supervised by pupil center was proposed, and iris boundary localization was speeded up by selecting the small block images and reducing the edge point number. Simulation results show that the algorithm is effective for the accurate and fast localization of irises in eye images.