Research On The Key Techniques Of License Plate Recognition System In Intelligent Park

With the fantastic spur in economy and rapid development of the owning amount of automobile, the requisition on the traffic control, safety management in the park area improves day by day. Using the License Plate Recognition(LPR) system to recognize he license plate, may realize the management automation and the intellectualization, which has the broad market prospect.The paper which had collected and summarized a lot of materials about LPR, has done some research in key Techniques of LPR system , including:(1) License plate location : One multi-level location method is proposed in this paper, which locates license many times, each time location reduces the analysis scope as far as possible, after many times locates the car license region finally. This method fully using the gray characteristic of license character and background, and its two-dimensional distribution rule, to search the license plate candidate region;Then in the candidate region which found, based on the edge examination, it search the boundary of the license plate. This method has the characteristic that it is not sensitive to the size of license plate and fake license plate, further more the time of location is short.(2) Character segmentation: After image pretreatment, the paper adopt the scan-line to eliminate disturbance factor on the top and bottom of license string by scanning twice. Single character is picked out by global partition with adjustment in local areas.(3) Character recognition: In regard to the deformed character and the different stroke of character, the paper adopt the improvable Hausdorff Distance to realize the template match algorithm based on the thin stroke of character. The experiment indicated that the method has many merits, such as stabilization of algorithm performance, highly recognition rate.A real-time license plate recognition system is designed and implemented. This system is applied in the platform of vehicles management to realize the automation and intellectualization of vehicles management in the park area.