Design And Implementation Of Fingerprint Feature Extraction System Based On FPGA

This dissertation used Nios Ⅱ and Quartus Ⅱ software produced by Altera, design and implement the fingerprint feature extraction system based on FPGA in EP2C35F672C6chip of cyclone Ⅱ series implementation. The hardware part of recognition system based on FPGA is designed and implemented by the Quartus Ⅱ software, and the software part of recognition system is to realize the recognition function of the system by the programming C code in the Nios Ⅱ.In the pre-processing aspects of image,Otsu’s image segmentation is implemented, then the orientation field and frequency field of the fingerprint image are calculated,the direction and frequency information are used in image enhancement, after that, we accomplished the binarization that based on slit method, Morphology-based thinning algorithm was refined.finally, we get a single pixel wide image of the fingerprint.Feature extraction can be done on the refinement image, the pseudo-features are canceled by using the ridge tracking technology, finally, we can get the set of points which can show the characterization of the fingerprint. Based on the above work,it will has a good foundation for the next matching processing.This dissertation used Nios Ⅱ soft-core processor implemented fingerprint feature extraction based on SOPC design and implementation. First, for the hardware,I designed and configured the fingerprint feature extraction system based on Nios Ⅱ soft-core CPU in the Quartus Ⅱ with SOPC Builder.I made the module for display by Verilog HDL,and embedded it in the former system to form an integrated system on programmable chip;Second, for the software,I edited application program based on fingerprint feature extraction and how to drive hardware,moreover,the program was compiled successfully,and transplanted to the Nios Ⅱ embedded development board to realize the system.