Research Of Finger Vein Image Acquisition And Recognition Technology Based On FPGA

The finger vein recognition technology is an important representative of the second generation of biological recognition technology,and the identity process is based on the vein texture features within the finger.Compared with the first generation of biological recognition technology,such as fingerprint identification,DNA identification,the finger vein recognition technology has three major characteristics,such as living body recognition,non-contacting,interior characteristic.With these characteristics,finger vein recognition technology is provided with high recognition accuracy,and easy to be accepted,more convenience and high security.There is a problem that the speed of finger vein registration and recognition in the finger vein identification system based on ARM,DSP and PC is relatively slow.FPGA has the advantage of fast processing speed.Aiming at this problem,this paper has carried on the research of the finger vein image acquisition and recognition technology based on the FPGA as the main control chip.The main achievements are presented as follows:(1)Based on FPGA as the main control chip,the paper has successfully developed finger vein image acquisition system which is composed of the finger vein image acquisition device,image acquisition module,image cache control module and LCD display control module.The finger vein collection and display system used in this paper is composed of near infrared light source module with the wavelength of 850 nm,MT9V034C12STM camera,the development board of VIP_Board Full FPGA,the TFT LCD display screen of 5.0 inch.The system is small in size,strong in real-time acquisition,clear in the finger vein image captured and has good stability.(2)The paper has put forward an algorithm of the finger vein feature extraction and matching recognition realized in FPGA through the analysis of the finger vein feature extraction and matching of mainstream recognition algorithm,and then the algorithm has been simulated in MATLAB to test and analyze.The paper has made the every 20 finger vein images of the right index finger of 50 people as the test object.The test results show that the finger vein feature extraction and recognition matching algorithm,which is proposed in the paper,is simple and effective,and the matching recognition accuracy reaches up to 98.3%(3)The key algorithms,such as the median filtering algorithm used to implement the finger vein image enhancement and filtering,finger vein image binarization,image erosion,image expansion used to achieve global feature extraction,of the finger vein image feature extraction and matching to identify has been realized by the Verilog HDL hardware language.This paper has developed a finger vein image acquisition display system based on FPGA.The experimental result shows that the system is small in size,strong in real-time acquisition,clear in the finger vein image captured and has good stability.This paper has realized some key algorithms of the finger vein recognition technology in FPGA.It has vital significance for the successful development of finger vein recognition system based on FPGA.