The Research On Palmprint Recognition Based On Line Feature

The biometrics technology can implement identity recognition used the physiological or acting characters of human being by computer. The application of the biometrics technology have two aspects:identity verification or identification. Identity verification is used to judge whether he or she is the one he or she said. Identity identification is used to tell us who is he or she.The palmprint recognition is a new biometrics technology,which use palm texture as biologic characters to recognise identity. For the great area,the abundant information,the steady characters,the palmprint can confirm the palm is from whose uniquely.Futhermore,the line features of palmprint are obvious in the bad quality images.So,in this dissertation,a process of identity recognition by the line features of online palmprint is described.The orientation and division are carried out on the orientation palmprint images. The contour of online palmprint are detected.Then,a new approach is used to extract feature points on the boundary. By this key points and their midpoint a coordinate is established for the orientation. A 128*128 center area of palmprint is divided up in this coordinate.The feature lines of the palmprint images are acquired and described. A 17*17 template is used to convert a grey image into binary image after enhancement. Then, Hilditch algorithm is used. An approach for getting rid of twig which brought forward by this dissertation is used.Finaly,a vector is defined for each feature line.Two steps are used for palmprint recognition in this dissertation.firstly,find out the disciplinal stylebooks which have the same number of feature line with the test stylebook;secondly,use the most adjacence algorithm to determine the class of the test stylebook.Experimental results show that the method of the dissertation is effective and available.