Biometric identification system using fingerprint and knuckle as multimodality features

Advancement of information technology coupled with the need for high security for passwords, the application of biometric as identification and recognition process serves a crucial purpose to avoid the risk to find or still. The texture pattern produces by hand geometry for authenticated access control it is the true substitute for passwords and identifiers in converse with the existing method biometric traits i.e. fingerprint and knuckle multimodal authentication system utilize line texture pattern to achieve actual personal identification. In this paper, we proposed the authentication system which uses multimodality feature extraction techniques with neural network as a classifier. Knuckle is a quite unique biometric trait having inherent skin pattern combined with fingerprint to implement multimodality. The performance of the system has been tested on biometric database that includes 16 subject's hands. The experimental result shows that multimodality features based authentication system has better authentication rate over single feature.