《Face and Fingerprint Biometrics fusion》.pdfVIP

Feature Level Fusion of Face and Fingerprint Biometrics A. Rattani, D. R. Kisku, M. Bicego, Member, IEEE and M. Tistarelli expected to provide more accurate results than the systems Abstract- The aim of this paper is to study the fusion at that integrate information at a later stage, because of the feature extraction level for face and fingerprint biometrics. The availability of more richer information. Since the feature set proposed approach is based on the fusion of the two traits by contains much richer information on the source data than the extracting independent feature pointsets from the two modalities, matching score or the output decision of a matcher, fusion at and making the two pointsets compatible for concatenation. Moreover, to handle the problem of curse of dimensionality, the the feature level is expected to provide better recognition feature pointsets are properly reduced in dimension. Different performances. feature reduction techniques are implemented, prior and after Fusion at matching score, rank and decision levels have the feature pointsets fusion, and the results are duly recorded. been extensively studied in the literature [3][4]. Despite the The fused feature pointset for the database and the query face abundance of research papers related to multimodal and fingerprint images are matched using techniques based on biometrics, fusion at feature level is a relatively understudied either the point pattern matching, or the Delaunay triangulation. Comparative experiments are conducted on chimeric and real problem. As a general comment,