The Harbor Weighing Bridge Management System Based On Vehicles' License Plates Recognition

As the important constituent of harbor information management system, the weighing bridge management system can effectively improve and enhance the level of the vehicles automation management in the harbor. Extremely, it has the market potential and the economic efficiency because it can speed up the circulation of the harbor goods traffic and reduce the harbor overhead charge. At present, harbor weighing bridge management is usually based on recognized vehicle' license information by artificially in our country. It lost large numbers of transported goods volume and money due to man-made factors. Some problems urgently need to be resolved, such as, how to decrease influence of man-made factors by adopting effective measures, how to assure security in the harbors. Therefore it has important practical significance to improve the automation's level of harbor weighing bridge management system.This article has conducted thorough research to the harbor weighing bridge automatic gathering technology taking computer vision system as method, comprehensively applying numeral technology such as picture processing, data communication, artificial intelligence and sensor and so on. Firstly, this paper has designed the function and the detection scheme to the harbor weighing bridge system on the basis of the literature search and study analysis in the investigation; Secondly, this paper has constructed the system platform of the harbor weighing bridge management in the harbor environment. It composed of the CCD camera, the vehicles' license plates recognition unit, the weighing bridge unit and industrial computer and so on.Next, this article introduces the technology of image pretreatment and localization, including image gradation, image enhancement, image edge examination, standardization and character segmentation and so on; After that, from the summary of the vehicles' license plates characteristic information, a vehicles' license plate localization method according to the image edge examination and the license plate region localization has designed. And the union template match optimization algorithm precisely withdraws the car license.Furthermore, after analyzing existing approaches of license plate binarization, the vehicles' license plate binarization based on the overall situation dynamic threshold value law is adopted; Using the improved Hough transform and image rotation, the rectification method can correct the image of license plate accurately; Then the method which combing of the connect region examination and least square method has been used to remove the frame and rivet of the vehicles' license plates; It designs an algorithm of vertical projection which based on the prior knowledge of the vehicles' license plates to achieve the separation, k can divide the image effectively when it having disturbance of complex background, character break or adhesion and so on. Finally, the method of combining the confidence analysis and the BP nerve network has been proposed to recognize the vehicles' license plates. The result has indicated that the recognition algorithm is satisfied the request of the good rapidity, the stability and the accuracy as well as real-time recognition.Finally, in order to facilitate the administrator to manage the vehicles of the harbor, the demand analysis, the concept structural design as well as the logical organization design has been carried on to construct the harbor weighing bridge database. In addition, this article has realized the algorithm and the interface of the harbor weighing bridge management software with the programming language of Visual C++. According to vehicles' license plate recognition unit, this system may realize addition, deletion, revise function about the vehicle type, cargo condition, customer information and charge information and so on. According to the actual situation, the massive experiments to the harbor weighing bridge management system have been carried on. The experimental results have indicated that, this system was stable, feasible, have a good practical value.