Study On Algorithms Of Palmprint Recognition

With the fast development of computer and internet, people's entertainment and economic activities have close relationship with computer and internet, and the whole society run more and more depending on network. The security mechanisms of confidential information protecting need to be strengthen continuously. Traditional security and authentication can not guarantee the safety of authentication. However, Biometrics based on personal biologic feature can solve the authentication problems efficiently and safely, which is concerned widely.Biometrics verify through physiological feature of human. Palmprint contain much texture information as the important part of human physiological characteristic and different people has different palmprint. It is also convenient to capture and easily accepted by people. For this reasons palmprint recognition can be used in personal identity authentication.This paper mainly studies on personal identity authentication based on palmprint recognition, especially on the preprocessing, feature extraction and matching. This paper extracts three kinds of preprocessed palmprint feature which has different property for palmprint recognition.Feature of principal lines spatial distribution: detect the principal lines through edge detecting technology, and then divide the image of palmprint into blocks with same dimension, compute the mean and variance as feature;Feature of palmprint direction: divide the image into sub-images; compute the max angle of projection and coordinate of each sub-image, which stand for the direction and location of the line on the palm;Feature of moment invariants: compute the moment invariants of each sub-image and the whole image as feature.This paper also presents matching strategy based on the three kinds of feature. The Experimental results illustrate the algorithm proposed reach good performance on recognition.