Combination approach to score level fusion for Multimodal Biometric system by using face and fingerprint

Abstract

Biometric System is used for person's recognition and identification for various applications. The Biometric system is unimodal and multimodal biometric system. Unimodal Biometric suffers from Noisy data, Intra class variation, non versality, spoofing etc. These drawbacks can remove by using Multimodal Biometric system. We developed the multimodal Biometric system by using Face and fingerprint Multimodalities. This system takes the advantage of individual Biometric System. This paper presents the fusion of face and fingerprint modalities at score level fusion. The system extracts the features and these features are then used for matching. Euclidean distance matcher is used for Face and Finger print modalities. Fingerprint recognition can be done with the help of minutiae matching and Gabor filter. The Face feature is extracted with the help of PCA (Principle Component Analysis) for dimensionality Reduction. Then the match scores are Normalized and sum score level fusion is used to develop the system. The proposed approach provides the better results. The Recognition Rate is increased and the error rate is decreased by with the help of this system.