Approach to Increase Accuracy of Multimodal Biometric System for Feature Level Fusion

Abstract

Biometric System are alternates to the traditional identification system. The Paper provides the multiple features based on the biometric system including Physiological and behaviouralchractersitics.like Fingerprints and iris which is used to identify the Fake and Genuine Users..In this paper we propose a Multimodal Biometric System for feature level fusion that combines the information to investigate the integration of fingerprints and Iris . This Proposed system extracts Gabor texture from the preprocessed fingerprints and Iris sample. The feature vectors attained from different methods are in different sizes and the features from equivalent image may be correlated. Therefore proposed the wavelet-based fusion techniques. Finally apply neural network’s Cascaded feed forward Back propagation Algorithm to Train Neurons for recognition.This approach is authenticated for their accuracy of Fingerprints virtual database fused with Iris virtual database of 16 users. The experimental results demonstrated that the proposed multimodal biometric system achieves a accuracy of 99.53% and with false rejection ratio (FRR) of = 1%