A Novel Accurate Algorithm to Ellipse Fitting for Iris Boundary Using Most Iris Edges

Ellipse fitting for iris boundary is an important step in eye gaze estimation and iris recognition applications. In most of the proposed algorithms only the edge image is used for ellipse fitting. In these algorithms, the undesired edges may lead to erroneous iris ellipse. In this paper, we propose a novel algorithm to ellipse fitting for iris boundary which uses gray and edge images simultaneously. In this algorithm, edges of eye region are computed and refined by some proposed constraints. Then, remainder segments merge together by an optimization process and the best ellipse is fitted for the merged segments. In this algorithm, most edges of iris boundary are used for ellipse fitting and hence, its accuracy is high. The performance of the method has been evaluated on real images. J Am Sci 2012;8(6):758-763]. (ISSN: 1545-1003). http://www.americanscience.org. 94