The Research Of Key Techniquse Of Finger Vein Recognition Based On FPGA

The people have higher requirements about security in twenty-first Century. In order to effectively protect the people's life, property and the security of the important information, state-of-the-art security system has already appeared imminent. Finger vein recognition has attracted worldwide attention since its birth. Finger vein recognition has the advantages of non contact, high security level, living recognition, internal features, et al. It has opened up new field of biological recognition, has become the second-generation biometric hotspot. FPGA (Field Programmable Gate Array) have made rapid progress in the past ten years, It came to the forefront of technology and has become the central processing device of digital system. The research of key techniques of finger vein recognition based on FPGA will certainly become a hot research topic in the field of biometric identification.In this paper, we have conducted in-depth study on many of the key technologies (such as image acquisition, enhancement, feature extraction and recognition, et al) of finger vein recognition and established a finger vein characteristics database. The main content of this paper is divided into the following section.(1) Finger vein image acquisition:Image acquisition is the key of the finger vein recognition system. The paper compares different near-infrared led, image sensor, filters and other auxiliary equipment, then select suitable devices throng experiments. Using near-infrared light irradiate the back side of the finger. On the other side with near-infrared light-sensitive imaging equipment to obtain image. The obtained image is the finger vein image.(2) The finger vein image processing:Owing to the finger vein image effect is poor, this paper analysis several classic image enhancement methods according the structure of the finger and finger vein feature. We combined with the multi-analyzing and direction of the valley detection to improve the effect of image enhancement.(3) Finger vein feature extraction, recognition:We analysis the characteristic points of the finger vein which can best embody the finer vein and extract endpoint and cross point of the finger vein network. The improved method of combination of rotational and translational template matching and Hausdorff distance authentication can improve the accuracy and speed of the finger vein recognition.(4) According to the research needs, the paper set up a small finger vein feature template database, as a template library for finger feature recognition.(5) FPGA system implementation:We establish the hardware and embedded software engineering according to the finger vein identify requirements. Finally, we realize some of the key technology of finger vein with FPGA.