A Personal Biometric Identification Technique based on Iris Recognition

Biometric methods designed for the identification of individuals are nowadays increasingly being used in the routine tasks like medical check-up, security etc. These powerful techniques are slowly replacing the less secure ID or password methods for user authentication. Iris recognition is especially attractive due to the high degree of entropy per unit area of the iris, as well as the stability of the iris texture patterns with age and health conditions. More recently iris recognition techniques have been proposed and implemented, most notably by Daugman. This technique is even more reliable than the solution based upon the traditional fingure print approach. It consists of five major steps i.e., iris acquisition, localization, normalization, feature extraction and matching. A number of groups have explored iris recognition algorithms and some systems have already been implemented and put into commercial practice by companies such as Iridian Technologies, Inc., whose system is based on the use of Daugman’s algorithm. In this work a new technique for the recognition of iris is proposed. Here, the iris images are acquired from the eye database of the University of Bath, England, and inner and the outer pupillary boundaries are detected for separating the pupil and the sclera from the iris portion. Then the localized iris image is normalized to get a rectangular portion. The correlation of every such normalized image is calculated with respect to one another and a proper threshold is defined accordingly. Finally applying suitable algorithm, the irises are matched with one another. This has shown promising results besides easy implementation and fast processing. Keywords— iris recognition, localization, normalization,