Biometric authentication by hand vein patterns

Since vein patterns of human are completely different each other, it is possible to identify the person by comparing the patterns. We propose the certification system that compares vein images for low-cost, high-speed and high-precision certification. The equipment for authentication consists of a near infrared light source and a monochrome CCD to produce contrast-enhanced images of the subcutaneous veins. We adopted phase only correlation and template matching as a recognition algorithm. We tested several noise-reduction filters, sharpness filters and histogram manipulations for best effort. As a result, we obtained high certification ratio in this system.