Abstracto

IRIS Authentication Based On AES Algorithm

R. Jubiya, M. Keirthi, M. Anupriya, A. Muthukumar

Cryptosystem is widely used in many information security application were the identification and verification are done by passwords, pin number etc which is easily cracked by others. So biometric cryptosystem is a powerful unique tool based on the anatomical and behavioral characteristics of the human beings in order to proof the authentication. Iris is used for generation of 128 bit binary key in order to prove the legitimate user who entering into the biometric system. This paper proposes extracting the multiple iris features from local iris image based on Independent Component Analysis (ICA) to reduce the dimensionality and to get accurate iris feature vector. It is clustered and its centroid value is converted into 128 bits iris key using K-Means clustering. The clustering is used to minimize the intra variation on the extracted iris feature vector. This paper also proposes biometric cryptosystem based on AES encryption and decryption to protect the confidential information by using iris as a biometric key. CRC is used to check whether there is no modification in the original data.