Palmprint And Palmprint Recognition System Based On SOPC

In this paper,a hand geometry and palm print recognition system is realized on FPGA.The palm image is acquired by CCD and preprocessed to obtain the palm edge contour.The palm contour is displayed by Sobel operator which is the most important operator in pixel image edge detection.In this system,a region of interest(ROI)selection algorithm for palm image is proposed,which can define and locate the valley point between the index finger and the middle finger,the valley point between the middle finger and the ring finger,the wrist point and the wrist center respectively.It also can determine the palm ROI area,and complete the image preprocessing process.In this paper,a morphome-based feature structure algorithm is proposed.By combining Sobel operator and morphological operator,the contrast of ridges in ROI area is further improved by adding sharpening,expansion,corrosion,histogram equalization,median filtering and other processes,so as to highlight the texture feature values generated by more detailed information as the palmgrain features in ROI area.The scale invariant feature transform(SIFT)algorithm is used to detect key points in the image and carry out image feature matching.In the recognition system,in order to compare feature codes quickly,hamming distance(HD)can be used to quickly compare a large number of feature codes,which can quickly realize the recognition of different hand geometry and palm prints.Finally,the whole recognition algorithm of hand geometry and palm print is transplanted to the embedded system based on FPGA,a hand geometry and palm print recognition system is realized on FPGA.Two core processors are designed on the FPGA system,one for image preprocessing,perform the operation of hand geometry contour extraction and segmentation of ROI region,the other one is used to extract and recognize palm print features,as external memory of processor,SRAM and FLASH are used to store the characteristic values of hand geometry and palm print.