An Improved Hashing Function for Human Authentication System

Abstract

Biometrics have the great advantage of recognition based on an intrinsic aspect of a human being and thus requiring the person to be authenticated for physical presentation. Unfortunately, biometrics suffer from some inherent limitation such as high false rejection when the system works at a low false acceptation rate. In this paper, near set are implemented to improve the Standard Secure Hash Function SHA-1 (ISHA-1) for strict multi-modal biometric image authentication system. The proposed system is composed of five phases, starting from feature extraction and selection phase, hashing computing that uses the ISHA-1 phase, embedding watermark phase, extraction and decryption watermark phase, and finally the authentication phase. Experimental results showed that the proposed algorithm guarantees the security assurance and reduces the time of implementation.