An Efficient Method for Finger-Knuckle-Print Recognition by Using the Information Fusion at Different Levels

Information fusion of various biometrics has attracted much attention in recent years. So in this paper we fused the information of biometrics in two different aspects. At the first, we investigate the information fusion in single modality, that is, the Finger-Knuckle-Print (FKP) biometric. FKP is one of the newest biometrics identifier which is recently used for personal identity authentication. For fusing the information of each FKP, two different representations of each image is used (Gray-Level intensity and its Gabor transform). On the other hand, two different subsets of feature vectors are extracted from each image. At the second stage, the information of each finger at two different fusion levels is fused: feature and matching score level. In fact this algorithm works as a kind of multi-modal method with a single biometric characteristic but multiple units. By fusing the information at different levels, the recognition rate can improve significantly. For example, by combining the information of four fingers, the recognition rate will be obtained 96.56% and 95.4% at feature and matching score levels, respectively. Poly-U Finger-Knuckle-Print database was used to examine the performance of the proposed method and the experimental results demonstrated the efficiency and effectiveness of this new biometric characteristic.