An Iris Recognition System to Enhance e-security



  bio-eye.doc (Size: 179 KB / Downloads: 0)

INTRODUCTION

In a race to improve security infrastructures faster than hackers and stealers can invent to penetrate passwords and firewalls, new technologies are being evolved to confirm or deny user authentication. Biometrics is a means of using parts of the human body as a kind of permanent password. Automatic recognition of persons has always been an attractive goal in informatics. As in all pattern recognition or classification problems, the key issue is the relation between interclass and intra-class variability: objects can be reliably classified only if the variability among different instances of a given class is less than the variability between different classes. In the example of face recognition, difficulties arise from the fact that the face is a changeable social organ displaying a variety of expressions, as well as being an active 3D object whose projected image varies with pose and viewing angle, illumination, accoutrements, and age. It has been noted that for images taken at least one year apart, even the best face recognition algorithms have error rates of 43% to 50%. For all of these reasons, iris patterns become interesting as an alternative approach to reliable visual recognition of persons.


Background and Related Work

Identification vs. Verification

It is important to distinguish whether a biometrics system is used to verify or identify a person. These are separate goals, and some biometrics systems are more appropriate for one than the other, though no biometric system is limited to one or the other. The most common use of biometrics is verification. As the name suggests, the biometric system verifies the user based on information provided by the user. For example, when X enters her/his user name and password, the biometric system then fetches the template for X. If there is a match, the system verifies that the user is in fact X. Identification seeks to determine who the subject is without information from the subject. Identification is complicated and resource-intensive because the system must perform a one-to-many comparison of images.

Biometric error analysis

All biometrics systems suffer from two forms of error: Form-1 is a false acceptance and Form-2 is a false rejection. Form-1 happens when the biometric system authenticates an impostor. Form-2 means that the system has rejected a valid user. A biometric system's accuracy is determined by combining the rates of false acceptance and rejection. Each error presents a unique administrative challenge. For instance, if we are protecting sensitive data with a biometric system, we have to tune the system to reduce the number of false acceptances. However, a system that's highly calibrated to reduce false acceptances may also increase false rejections, resulting in more help desk calls and administrator intervention. A poorly created enrollment template can compound false acceptance and rejection.


The iris features and process

The iris has many features that can be used to distinguish one iris from another. One of the primary visible characteristics is the trabecular meshwork, a tissue which gives the appearance of dividing the iris in a radial fashion. The fact that the iris is well protected from environment and it remains in a stable form from about the age of one until death, unlike other biometrics such as fingerprints, the likelihood of damage and/or abrasion is minimal. The use of glasses or contact lenses has little effect on the representation of the iris and hence does not interfere with the recognition technology.


Overview of our approach:
In our method we use a two level secured Iris recognition system. In the first level we use the diameters of the pupil and iris to verify for matching. The average inner and outer diameters of iris are taken from different samples of same eye image with different features (smiling, crying etc). In the second level, the sharp variations occurring in the structure of iris are used for matching. Based upon these sharp variations, a binary sequence is developed and further coded into a unique ID for every user.