As an important part of the license plate recognition system, the recognition speed and efficiency of plate number character recognition decide the performance of license plate recognition system directly. At present, most of the existing vehicle license plate character recognition systems are based on software and it is difficult to achieve the requirements of real-time and recognition speed.In this paper, by analyzing the merits and disadvantages of current license plate character recognition methods, the FPGA logic resources and implements are used to achieve the hardware architecture design based on2DPCA method. Then the modules including image scaling, characteristic matrix distance calculation module, the minimum distance calculation module and its realization process in the architecture design are optimized, and at last taking the hardware architecture design to Altera DE270FPGA hardware platform and validated ulteriorly, the verification results show that the hardware architecture based on2dpca license plate character recognition algorithm is effective and feasible. |