A novel cryptosystem based on iris key generation

Abstract

Biometric cryptography is a technique using biometric features to encrypt data, which can improve the security of the encrypted data and overcome the shortcomings of the traditional cryptography. This paper proposes a novel biometric cryptosystem based on the most accurate biometric feature - iris. In encryption phase, a quantified 256-dimension textural feature vector is firstly extracted from the preprocessed iris image using a set of 2-D Gabor filters. At the same time, an error-correct-code (ECC) is generated using Reed-Solomon algorithm. Then the feature vector is translated to a cipher key using Hash function. Some general encryption algorithms use this cipher key to encrypt the secret information. In decryption phase, a feature vector extracted from the input iris is firstly corrected using the ECC. Then it is translated to the cipher key using the same Hash function. Finally, the corresponding general decryption algorithms use the key to decrypt the information. Experimental results demonstrate the feasibility of the proposed system.