Research On Identity Recognition Technology Based On The Back Of Hand Vein

With the rapid development of information technology and the increase of the economic activity and people's social activities,identity has become more important.Traditional authentication(Password identification,password authentication,etc)can't indentify the user itself.Its security is poor,and it can't meet the needs of today's society for identification.Dorsal hand vein recognition technology,as one of the biometric identification technology,brings hope for solving this kind of security problems.In this paper,the machine vision recognition technology is used to collect the hand vein image,carrying on the pretreatment to the hand-back vein image and studying the recognition method on the basic of collecting image.The paper designs a hang-back vein image acquisition system based on machine vision.In order to obtain high-quality hand vein image less disturbed by the outside environment,this paper selects the appropriate light source,the light barrier,image sensors and lens with other components.In order to extract the data features of the dorsal vein structure effectively,the image need the necessary pretreatment,including the correction of the back of the hand,image segmentation,noise processing,image enhancement,refinement of vein structure.Meanwhile,the normalization processing of the image is performed,and the effective area of the image is extracted.According to the charateristics of dorsal vascular veins,a variety of pretreatment methods,combining with MATLAB simulation to get the corresponding experimental results,which has laid a good foundation.On the basis of image preprocessing,the features of the dorsal vein were extracted.According to the requirement of hand vein recognition,A recognition model combining the back vein feature extraction algorithm and the BP neural network algorithm was established.In the hand vein feature extraction algorithm,the principal component analysis-linear discriminant analysis(PCA-LDA)and thenuclear two-dimensional principal component analysis-two-dimensional linear discriminant analysis(K2DPCA-2DLDA)are used to extract the dorsal vein feature.The extracted component variables are trained as inputs to the BP neural network,and the network parameters are fixed.After the improvement of the neural network,the trained neural network is used to identify the dorsal vein.The simulation results show that the algorithm based on K2DPCA-2DLDA-BP model has improved the accuracy and real-time,and has achieved the purpose of research.In the paper,the back of hand vein recognition technology of the study is put into in the test system,though software design of the system,which implements the hand-back vein indentification and verifys the effectiveness and the practicability.