A New Personal Authentication Based On Hand Back

System and information security is of increasing importance in E-world, which brings about great challenges to personal identification field. As a novel identification technology, biometrics, which uses the inherent traits of human body to automatically identify individuals, has emerged as the times required. Comparing with the traditional personal authentication methods which always utilize codes or IC card, biometrics cannot be lost, forgotten, impost and copy, and therefore is more safe, reliable and efficient.With the rapid development of biometrics, people are continually looking for new traits to identification. Up to now, many traits in human body have been applied, such as fingerprint, palmprint, iris, face, hand shape and signature, etc. In the thesis, we propose a novel personal identification technology base on hand back and develop a prototype of hand back based Identification in this paper. The thesis aims to solve the problems in the information acquisition and processing, feature extraction and classification in the hand back based Identification. The main contents are as follows:(1) Manufacturing a hand back acquisition device for hand back based identification, which can reduce the rotation and displacement of hand during the hand back images are captured, and build up a hand back image database which contains 600 images.(2) Researching a novel 2DPCA algorithm which extracts both horizontal and vertical is also proposed to choose and classify features.(3) Base on the correction and improvement of Chien's NFS, we advance a novel NFS. The proposed method removes redundancies and uncertain points, and addresses the original point problem in Chien's NFS.(4) We do experiments by applying the novel 2DPCA algorithm and the proposed NFS based on our hand back image database. The experimental results show that the hand back based identification can effectively identify individuals, and is of great perspective.