Research On Palm Vein Image Acquisition And Thining Technology

Palm vein recognition technology proposed in recent years is a new biometric technology. Palm vein recognition put a person's palm vein as an object to be recognized,which can recognize different individuals through the acquisition, enhancement,binarization,thinning,feature extraction, matching.Palm vein recognition technology proposed in recent years is a new biometric technology. Palm vein recognition is a person's palm vein recognition as an object, through the acquisition,enhancement,binarization,thinning,feature extraction, matching a series of treatment processes to identify different individuals technologies.It widely used in these fields such as government, military, banking, social welfare and security, electronic commerce, security, defense and so on, so it has enormous advantages and vast market potential for applications as the latest and most secure biometric technology .The paper puts emphasis on researching the technology of acquisiting ,extracting and thinning the palm vein.To achieve the goal that acquires a complete one-pixel framework of palm venous network, this paper discusses the improvement that has been made in the extraction and thinning algorithm of palm venous image based on the analysis, conclusion and summarization of the existed algorithm through obtaining the original infrared palm venous image. The main work is represented as follows:The first is to obtain the high qualified infrared palm vein image. After the analysis of the basic principle of the palm vein image, equipments consistent with the requirement of experiment are chosen and the collection device is produced through the comparison of different light source devices, imaging devices, filters and other auxiliary equipment. The experiment has proved the availability of the palm vein image acquired.In the process of extracting the vein lines of the palm from the acquired palm vein image, an improved binarization method based on the integration of Niblack algorithm and block algorithm is proposed to solve the problems of noise and disconnected lines occurring in the process of palm vein image binarization based on the Niblack algorithm. Then the denoise algorithm of mathematical morphology and connected component label is applied according to the characteristics of the palm vein in the following binarization denoise algorithm. The experiment has proved the validity of this method.Then it comes to get the complete one-pixel framework of palm vein through the analysis and comparison of the advantages and shortages of the classical thinning algorithm and the optimization of the burning and thinning algorithm.The experimental result shows that the improved algorithm is superior to the existing algorithm through the validation of the improved algorithm and the comparison and analysis of the above two kinds of the algorithms. The acquired palm vein image is matched well with the distribution of the original palm vein lines completely which satisfies the need of the following feature extraction and matching.