Efficient iris localization and recognition

Iris localization is an important step in iris recognition systems; all the subsequent steps, iris normalization, feature extraction and matching, depend on its accuracy. Traditional iris localization methods often involve an exhaustive search of a three-dimensional parameter space, which is a time consuming process. This paper presents a coarse-to-fine algorithm to address the computational cost problem, while achieving an acceptable accuracy. The iris gray image is transformed to a binary image using an adaptive threshold obtained from analyzing the image intensity histogram. Morphological processing is then used to extract an initial center point, which is considered as the initial center for both pupil and iris boundaries. Finally, a refinement step is made using an integro-differential operator to get the final iris and pupil centers and radii. This system proves to be robust against occlusions and intensity variations.