New Algorithm of Iris Localization

In order to enhance the accuracy of the iris localization, a new algorithm is proposed by combining the merits of wavelet analysis and Hough transform. For the inner edge localization, the wavelet analysis is applied to the iris image to remove the noise availably. Then using the Hough transform locates the iris inner edge accurately. Thereby the invalid search brought by the noise is avoided. For the outer edge localization, the data of iris inner localization are used to wipe off noise so as to get effective edge information. Based on this method, an improved Hough transform is used to define the iris outer edge fast and accurately. The results prove that this method not only improves the speed and the precision of iris localization, but also avoids the magnitude of compute.