10 Iris Recognition System Using Support Vector Machines

In the modern world, a reliable personal identification infrastructure is required to control the access in order to secure areas or materials. Conventional methods of recognizing the identity of a person by using passwords or cards are not altogether reliable, because they can be forgotten, stolen, disclosable, or transferable (Zhang, 2000). Biometric technology, which is based on physical and behavioral features of human body such as face, fingerprint, hand shapes, iris, palmprint, keystroke, signature and voice, (Lim et al., 2001, Zhang, 2000, Zhu et al., 1999) has now been considered as an alternative to existing systems in a great deal of application domains such as bank Automatic Teller Machines (ATM), telecommunication, internet security and airport security. Each biometric technology has its own advantages and disadvantages based on their usability and security. Among the various traits, iris recognition has attracted a lot of attention. Iris is an internal (yet externally visible) organ of the eye, which is well protected from the environment and its patterns are apparently stable throughout the life. The iris consists of variable sized hole called pupil. The average diameter of the iris is 12 mm, and the pupil size can vary from 10% to 80% of the iris diameter. It has the great mathematical advantage that its pattern variability amongst people is enormous (Daugman, 2002). The number of features in human iris is large. Its complex pattern can contain many distinctive features such as arching, ligaments, furrows, ridges, crypts, rings, corona, freckles and zigzag collarette (Wildes, 1999, Daugman, 2002) for personal identification. Fig. 1 is an example of human iris. That is because every iris has fine unique texture and does not change over time. In addition, iris pattern can have up to 249 independent degrees of freedom. Because of high randomness in the iris pattern, it has made the technique more robust and it is very difficult to deceive an iris pattern (Daugman, 2003). Unlike other biometric traits, iris recognition is the most accurate and non–invasive biometric for secure authentication and positive identification. This proposed system use a publicly availably library for iris recognition written in MATLAB (Masek, 2003). Due to the advantages of iris recognition systems which offer reliable and effective security in the present day, this research proposed the use of iris-based as verification system system to identify the person’s identity. This research work adopts Support Vector Machines (SVMs) as pattern classification techniques which are based on iris code model which the feature vector size is transformed to one-dimension vector which reduces to 1 x 480 by using averaging techniques (each segment is divided by 20) contains the average value to