Palmprint Recognition Based on PFI and Fuzzy Logic

Palmprint is one of the most reliable features in personal identification because of its stability and uniqueness. This paper proposed a novel palmprint recognition algorithm based on principal lines: a new irregular geometrical shape was employed to get valid palmprint region, which decreased the influence of large noises; not only structure feature, but also intension information were included in the final extracted principal lines, which provides much sufficient clues for palmprint recognition; the probability feature image (PFI) was used in order to suppress random noises in feature image; features from several training samples were merged into one template, which guaranteed the integrity of feature; fuzzy logic was employed in matching algorithm. The high recognition rates obtained in the experiments show the efficiency of proposed algorithm.