License Plate Recognition Algorithm Transplantation And Optimization Based On DM6437

License Plate Recognition system is one of the most important parts in Intelligent Transport System. It is frequently used in various situations, such as vehicle flow control, access control, and electronic toll collection. People is paying more and more attention to the research of related technologies, since it can improve the efficiency of traffic management and reduce labor cost through the application of automatic management。The paper’s main purpose is to transplante License Plate Recognition algorithm into embedded DSP platform, License plate recognition technology and DSP to achieve the integration and testing of the video stream, This paper firstly discussed the hardware development platform and studied the algorithm theory. Then start to transplant and optimize the algorithm. Optimization based on DM6437, including floating-point operations, storage, memory optimization, inline functions and applications such as histograms and structural optimization of algorithms. And then did some test on the algorithm which was transplanted into DSP.After analysis and verification of the results, we come to a conclusion that the License Plate Recognition algorithm transplanted in this paper has very good effect on DSP platform. Both the detection speed and accuracy have met the application requirements.