Design of Multimodal Biometrics Authentication using Feature Extraction and Fusion

Multimodal biometric can overcome the limitation possessed by single biometric trait and give better classification accuracy. The present work proposes an authentication system with the fingerprint, face and iris multimodal biometric system based on fusion at the feature level. The performances of fingerprint, face and iris recognition can be enhanced using a proposed feature selection method to take an optimal subset of features. Fingerprints are the most popular and studied biometric features. Their stability and uniqueness makes the fingerprint identification system extremely reliable and useful for security applications. An optimized fingering print algorithm is used to extract the ridge count, ridge length, and ridge curvature direction features from a fingerprint. Iris is also a unique biometric feature, advanced iris and sclera algorithm is used to extract the iris and sclera features of the eye and Eigen face feature extraction algorithm is used to extract the face feature extraction. The results indicate that the proposed feature selection method is able to improve the classification accuracy in terms of total error rate. Index Terms Biometrics, Multimodal, Face, Fingerprint, Iris and Fusion.