Research On Iris Recognition Algorithm

In age of electronic information, a more convenient and effective identity verification technology is needed. Biological recognition technology makes use of the inherent physiological or behavioural characteristics of human beings to identify individuals. Due to the unique, stable and live physiological properties of the iris, non-invasive to users, and most stable, reliable for identification in practical applications of iris-based system, iris recognition is receiving extensive attention and becoming an active topic in Biological recognition technology.In this dissertation, the particular advantage of iris for recognition is discussed and summarized. And then some of the key issues related to iris recognition technology are investigated, including iris acquisition, iris pre-processing, iris feature extraction and matching. At the basics of analyzing the mechanism of eyeball movement, saccade movement of eyeball is used for live iris detection. Based on the grey scale threshold and edge tracing the pupil is located. The outer circle is detected using Canny operators together with Hough transform. In order to remove the effects caused by displacement and zoom of iris image, the iris image is transformed from Cartesian coordinates to polar coordinates. For reducing the influence of unsymmetrical illumination, the iris image is transformed by histogram equalization.One of the cores of iris recognition technology is the iris feature extraction. The method of wavelet packets decompositions originating from wavelet transform is more accurate in signal analysis, with the predominance of analyzing details information of high frequency. Except that, singular value, as an excellent algebraic texture, can describe the substantial character of image signal effectively. The method, based on wavelet packets decompositions and singular value decompositions, is applied to extract iris feature in this dissertation. Euclidean Distance is adopted to perform pattern matching between two irises.Experiment results for our iris database prove that the approach is effective.