Application Research of MD5 Algorithm in LSB Watermarking

Abstract

As malicious attacks greatly threaten the security and reliability of biometric systems, Authentication of biometric data is very important. In this paper multimodal authentication based on watermarking is proposed to address this problem. Face features of an individual are embedded into a fingerprint image which works as data credibility token and secondary authentication source. In first stage of authentication, validity of extracted patterns is checked with the input biometric data. Due to the specific characteristics of face feature, the face feature detection based classification strategies are used for reliable watermark verification. If token authentication is successful, the face features works as  additional identity information to facilitate sub sequential biometric authentication.

In this framework, one critical issue is to guarantee the authentication to images and capacity of watermark while preserving the discriminating features of host fingerprints. Hence a MD5 algorithm is used to generate a key of face image and LSB watermarking approach is proposed to distribute watermark energy of fingerprints. Experimental results shows both watermarking and biometric authentication performance and demonstrate the effectiveness of this work.