Research On Palm-Dorsa Vein Recognition Method

With the advent of the information age, information security is particularly important. How to identify someone accurately and protect information security is a key issue should be solved in today's information day. The identity authentication technology based on biological characteristics has showed a great of advantages because of the use of inherent physiology and behavior characteristics of human. Among them, the palm-dorsa vein recognition is a new non-contact biometric technology which received a lot of attention in recent years.The palm-dorsa vein recognition is a method of recognition by analyzing the feature of the vein of palm-dorsa. In this paper, I analyzed the theoretical basis of vein recognition and its advantages to other biometric recognition. In the specific research, I summarized algorithms related to vein recognition about effective region, image segmentation, feature extraction and feature match.In the vein image preprocess module, first, the effective region of palm-dorsa vein had been positioned, and then the picture is processed, such as image enhancement and image segmentation. The main method of enhancement included normalized mode, median filter, low pass filter. It is proposed that a method in image segmentation base on threshold graph. Experimental results proved it stable and feasible. After segmentation, it is necessary to reduce the noise and refine the picture to get the final picture.In the feature extraction and matching module, this paper made a lot of tests and new methods, such as a new way which can be used in the palm-dorsa vein recognition is to directly project the image matrix based on Two-Dimensional FLD, a method on the structural feature of vein, an algorithm based on the structure feature of vein. In the moment's way, it is proposed that seven square definition and method of template matching. At the same time, they had achieved well recognition effect. The Two-Dimensional FLD experimental results on 57 samples in the database showed that the recognition rate reaches 96.49%, which is an effective way of palm-dorsa vein recognition. Finally, it concluded that some ideas and the future research about palm-dorsa vein recognition.