The present invention provides a
hand vein recognition method based on
curve matching. The method comprises the steps of firstly scanning the
vein branch curves according to the
venous line endpoints and the intersection points, then extracting the characteristics of each curve, such as the curvature, the length, etc., and finally, realizing the reliable recognition of the hand veins by combining the coarse screening and the
curve matching. Under the premise of enhancing the anti-interference ability of the characteristics, the inter-cluster distinguishing ability of the characteristics is improved, and the reliable recognition of the hand veins is realized. The
hand vein recognition method based on
curve matching of the present invention solves the problems that the
sample rotation and translation influences the recognition, the
algorithm matching speed is accelerated, etc., and the efficiency of the method accords with the requirements of a mode
recognition system. The
hand vein recognition method based on curve matching of the present invention enables the hand
vein recognition performance to be improved effectively, and is widely used in an intelligent
access control system.