Positioning System Based On Iris Capture

With the development of science and technology, the iris recognition technology shows great superiority as an emerging biometric technology in social security, financial security and aviation safety aspects. Iris image quality evaluation is a significant step of iris recognition, because the image quality is crucial to recognition effect. Iris location is very important to iris recognition system and the time is about half of the identification process. So the accurate and speed of location algorithm affect the recognition rate and efficiency of iris recognition system. In the paper, some algorithms of iris quality evaluation and iris location are analyzed. Due to the characters of iris capturing system, a system with capturing eye image and locating iris boundary is designed. The research work is done mainly as follows:1. At present, the current algorithms of image quality assessment are still limited. An improved method is proposed to realize iris image quality assessment. It includes overall assessment of iris image by intensity, determining the location of bright spots by calculating the distance between the pupil's center and spots, evaluating definition of iris image through total bounded variation and estimation of effective region by linear fitting approach.2. Due to the influence of light spots and the difficulty to select a threshold satisfied for edge extraction in exacting iris edge image, a method of edge detection is proposed based on local extreme value of image's gradient. Based on this method, points of iris boundaries can be detected accurately, and the threshold selection problem needn't be considered. The effectiveness of the edge extraction is improved.3. In order to improve the real-time performance of iris localization algorithm, a fast iris location method is proposed based on regional of interest. Iris boundaries can be located quickly by using the method.4. A software is designed based on OpenCV and VS2008. This system not only can capture iris image, but also can evaluate the iris image's quality. Iris image with high quality can be obtained and then can be used for extracting iris texture feature and pattern recognition.All the algorithms are tested in Matlab 7.0 and the results show that the methods are feasible. The performance of iris processing system based on OpenCV and VS2008 are satisfied.