Technique Study Of Hand Vein Image Recognition

In company with the social developing and technology improving, biological feature recognition is getting much closer to people's daily life. Fingerprint recognition is considered as the comparative maturity technology in biological feature recognition whose algorithm performance has applied in practical products. However, Fingerprint recognition has defects, such as overwet, overdrying or cruel skin of finger or contamination of finger would contribute to the results. So we bring up a proposal of identification based on venous recognition back of hand. Body interior characteristic recognition will reduce the external influence and prevent falsification.The dissertation discussed the system developing of venous recognition back of hand.First, it is discussed that the background, purpose and the meaning of the theme. The venous recognition back of hand has many merits such as living body recognition, interior characteristic, non-contacting and high safe class so the research is meaningful and feasible. In analyzing the evolution of the technology home and abroad we got the clear goal of the research. At the same time, this part gives out the keystone and difficulty of the systemic frame structure.Second, it is discussed that the hardware design of venous imaging. It is studied that the principle of IR picture picking and choose MICRON's MT9V011P11ST as the color image sensor. In comparing several communication interface circuits, USB interface was considered the best choice. CYPRESS'S CY7C68013 was chose as the USB control unit. The hardware interface circuit and the main code in communication were given. The device was used for collecting experiment. The results were studied and the reason was listed.Third, raw venous picture was processed before characteristic abstracting. The picture is preprocessed first, such as image enhancement and the normalization of size and grayscale. It is proposed that a method in image segmentation base on threshold graph, whose algorithm set threshold on every pixel of the picture and then split together with the border. Experimental results proved it stable and feasible. After segmentation, it is necessary to reduce the noise and refine the picture to get the final venous picture.