Application of metaheuristic for optimization of iris Image segmentation by using evaluation Hough Transform and methods Daugman

The biometric domain is very important for multinational corporations and economic companies all over the field of research and innovation and the protection of personal confidential information. Thus, governments invest millions of dollars in the protection and the identification of citizens' personal data and use different types of biometric data modalities (iris, fingerprints, faces, etc.) for the recognition of the individual. The iris modality requires special and adequate treatment so that the biometric recognition system can recognize the person and this by going through several stages: image acquisition, segmentation, normalization, feature extraction and finally matching. The segmentation is a delicate step for the iris localization. The process detects the iris by two concentric circles, the first circle for the pupil region and the second circle for the iris region. The two most widely used methods in the literature are the John Daugman method (Integro-Differential Operator) [7] and the Hough Transform circular [8]. The method of John Daugman detects the iris region and the pupil as well as the arches of the upper and lower eyelids. The Hough circular transform makes it possible to obtain the coordinates of the radius, the center of the pupil and the iris. The main objective is to obtain the essential information, which is the size of the pupil radius and the iris radius. And that makes it possible to crop the radius rl of the pupil and the radius r2 of the iris to have a good segmentation. Experiments were made on the following iris data: Casia-iris-interval v4 [21], Ubiris vl [23], MMU 1, MMU 2 [22] and the SDULMA-HCM iris database [24].