3D iris structure impact on iris recognition

The accuracy of iris recognition systems depends on several factors such as the quality of data capture, camera angle, pupil dilation. In standoff iris recognition systems, a new biometrics research area, corneal refraction of light, depth of field blur, and the limbus effect are previously addressed as degradation factors for the iris recognition. Since iris consists of dilator muscles to contract the pupil, it has a three-dimensional (3D) structure. Therefore, 3D structure of the iris cause additional significant performance degradation for the standoff iris recognition systems. However, in previous studies, 3D structure of the iris is ignored by assuming two-dimensional iris plane. In this paper, we investigate the effect of the 3D iris structure on iris recognition by using eye models with different iris models. Based on the result, 3D iris structure effect increases the hamming distance as the angle between frontal and off-angle iris images increases. Therefore, three-dimensional iris texture must be taken into consideration to obtain accurate results in the standoff iris recognition.