Study On The Recognition Of Hand Vein Pattern

With the development of communication and Internet industry, human society has be in an era of information. And the importance of information security has been rised to an unprecedented level. In order to ensure the safety of information, the first important problem is the identification. For this reason researchers found biometrics, a safe, reliable and convenient identity recognition technology. Biometrics refers to the identification of humans by their characteristics (e.g fingerprint, iris, DNA) or traits (e.g gait, voice). The future biometrics will have more broad market prospect, and due to its importance, it attracts researchers’more and more attention, becomes an important research focus in the field of pattern recognition.Hand vein recognition is a new biometric technology, which develops rapidly in recent years. It has many excellent features such as uniqueness, stability, hard to copy and forgery, and easy to fuse other hand features such as hand geometry, palmprint, fingerprint to compose a multimodal biometric system. Hand vein recognition is not only a significant biometric technology, but also an important supplement to the current biometric technologies. Therefore it is believable that hand vein recognition will have great potential applications. However, hand vein recognition is still in the developing stage now. Under this background, this paper focuses on the solutions of some key problems in hand vein recognition.The main contents are as follows:①hand vein image acquisition and pre-procesing, including designed image collection device, builded a sample library of2040images, image denoising, ROI extracting, and vein segmentation.2)a hand vein recognition method based on line feature is proposed which reflects the change of orientation projection. In order to obtain better performance, we have improve the method by adding weight information and the final identification rate is above94%.3) a hand vein recognition method based on texture feature is proposed. The principle of LBP and PLBP is introduced in this paper. Finally a novel approach to extract multi-scale LBP features of hand vein images is proposed.4) an access control system based on hand vein recognition is designed. This system adopted Client/Server model, and has enroll, identify, and access control switch functions, and achieved the design requirement.