Recognition of Bimodal Biometric System using Transformation Techniques

A wide variety of systems requires reliable personal recognition schemes to confirm or determine the identity of an individual requesting their services. In this paper, recognition of Iris and Fingerprint provides easiest way of security and faster processing. It ensures data security and protection of system from unauthorized users as well. In first part, Iris template is generated from Iris image. Iris features are generated by applying Discrete Wavelet Transformation (DWT) and Discrete Cosine Transformation (DCT) on Iris Template. Second part focus on fingerprints; the Fingerprint is preprocessed to get Region of Interest. Using DWT and Fast Fourier Transformation (FFT), features of Fingerprints are obtained. Final feature set is generated using concatenation. The final feature is compared with stored database using Euclidean distance matching to obtain exact match depending upon threshold value. We check False Acceptance Rate (FAR) and False Rejection Rate (FRR) at different threshold level.