With rapid economy development, secure and reliable identification techniques are needed in more and more fields, such as automotive, banks and financial departments, and door lock system. Fingerprint identification is an effective identification. A research on the fast-operating, portable, economic fingerprint acquisition and processing system is very useful, due to its practical value.The advantage of FPGA's parallel computing and being easily integrated to the other systems can give a firm support to fingerprint usage. Based on the Xilinx FPGA, a fingerprint acquisition system is designed. Firstly, the fingerprint acquisition technique and Xilinx FPGA developing environment are briefly introduced. Then, the system design architecture is proposed. The structure and working principal of the fingerprint sensor MBF200 are discussed. Utilizing the SPI interface provided by MBF200, a fingerprint acquisition board is designed, fabricated and debugged. The functional blocks are designed and simulated in FPGA developing environment. Lastly, data is transferred to PC by RS-232. By using Matlab processing, better fingerprint images are got. In the paper the fingerprint preprocessing algorithm is also researched. In order to get high definition, the preprocess, such as fingerprint image division and enhancement, Binaryzation, and thinning is applied. For the most important part of preprocess fingerprint image enhancement, directional filtering algorithm is adopted by utilizing the self-characteristics. Directional diagram calculation and directional filter design method are given in the paper. Algorithm simulation is carried out using Matlab. The result shows the algorithm can get the expected function.The system test result shows that hardware system is operating well and good fingerprint image can be collected. After the preprocess the image becomes better. The specifications are met in major. It made a firm basis for the future study of fingerprint characteristics collecting and identification accuracy improving. |