Rotation Invariant Approach of SURF for Unimodal Biometric System
The people all have unique physiological characteristics that distinguish one from other people. Biometrics uses these unique characteristics (or identifiers) to ascertain and verify people identity. Fingerprint recognition involves the location and determination of the unique characteristics of the fingerprint. The fingerprint is composed of various 'Ridges' and 'Valleys', which form the basis for the loops, arches and swirls on their fingertip. The ridges and valleys contain different kinds of breaks and discontinuities. In this paper, the authors present improved performance evaluation parameters like Accuracy, FAR, FRR and EER for thumbprint (Uni-modal) biometric system using SURF approach. Here, they also compared the proposed approach with existing methods e.g. STFT, SIFT, etc.