Research On Palmprint Identification System

As a network identification and information age essential identification means,biometric identification technology has become a popular research frontier at home and abroad.Palmprint recognition,as a new biometric technology,has the advantages of high reliability,fast identification,low cost,easy to use and high degree of acceptability,has attracted wide attention and attention,and has high research value and extensive Application prospects.This paper studies the related technology of the palm image image recognition system and designs a palmprint identification system.This article has done the following work:1.The preprocessing technique of palmprint image is studied,including the grayation,binarization,localization of the palmprint image,the cropping of the region of interest and the normalization of the size.Through the research and analysis,this paper improves the palmprint image localization and segmentation algorithm,and proposes a boundary tracking method based on the eight-direction chain code of the palm contour to extract the contour feature points,and then realize the image segmentation segmentation based on the contour feature points and detailed The principle of the algorithm is introduced.Finally,the comparison experiment is carried out,and the experimental results are analyzed.The experimental results show that the improved algorithm proposed in this paper is better than the disk-based algorithm,and the accuracy rate is more than 95%,and the time is only 0.3s,to achieve the improvement effect.2.This paper studies various kinds of palmprint image recognition techniques and improves them according to the system requirements designed in this paper,and proposes a recognition algorithm combining two-dimensional Gabor wavelet transform and Fisher's palm,The principle of the algorithm is described in detail.Finally,the comparison experiment is carried out,and the experimental results are analyzed.The experimental results show that the palmprint recognition method proposed in this paper is better than the two-dimensional Gabor wavelet transform method and Fisher palm method,and the recognition rate is more than 95%,and the use of these two methods to identify the lack of identification,to achieve the improvement effect.3.Designed a palm image image recognition system,to achieve the user registration,user delete and user identification function.Then,the performance test of the system,the test results show that the system has a good recognition rate(less than 5%error rate),and better identification efficiency(time required for about 0.5s).