An enhanced iris recognition and authentication system using energy measure

Abstract

In order to fight identity fraud, the use of a reliable personal identifier has become a necessity. Using Personal Identification Number (PIN) or a password is no longer secure enough to identify an individual. Iris recognition is considered to be one of the best and accurate form of biometric measurements compared to others, it has become an interesting research area. Iris recognition and authentication has a major issue in its code generation and verification accuracy, in order to enhance the authentication process, a binary bit sequence of iris is generated, which contain several vital information that is used to calculate the Mean Energy and Maximum Energy that goes into the eye with an adopted Threshold Value. The information generated can further be used to find out different eye ailments. An iris is obtained using a predefined iris image which is scanned through eight (8) different stages and wavelet packet decomposition is used to generate 64 wavelet packages bit iris code so as to match the iris codes with Hamming distance criteria and evaluate different energy values. The system showed 98% True Acceptance Rate and 1% False Rejection Rate and this is because some of the irises weren’t properly captured during iris acquisition phase. The system is implemented using UBIRIS v.1 Database.

Keywords: Local Image Properties, Authentication Enhancement, Iris Authentication, Local Image, Iris Recognition, Binary Bit Sequence