Security

Multimodal Personal Authentication with Fingerprint, Speech and Teeth Traits Using SVM Classifier

Download Now Date Added: May 2012
Format: PDF

Multimodal biometrics systems are becoming increasingly efficient over the unimodal system, especially for the securing mobile devices like PDA, PC tablets and, etc. In this paper, the authors propose a novel tri-modal biometric recognition technique using teeth, fingerprint and voice as biometric traits. The matching scores of the individual traits are classified using support vector machine. The experiments were conducted over a database collected from 20 individuals with multiple instances of all the three traits. The performance analysis of the fusion techniques revealed that the equal error rates of 1.44%, 1.88% and 3.06% for the support vector machine, weighted summation and K-NN Classifier respectively.