The Design Of The New License Plate Positioning System

The technology of License plate recognition concerns digital image processing, pattern recognition, and other techniques, accurately and automatically identifies the information of vehicle license plate in the image, achieving the vehicle automation management. License plate recognition is the key technology of intelligent transportation systems and is widely used in traffic flux detect, electronic police system, vehicle manage of community, automatic charge of highway and the monitor of violate the rules and regulations.The technology of License plate recognition includes three parts:License plate location, characters segmentation and characters recognition. The license plate location is the critical step to achieve license Plate recognition, which is one of the important research subjects in intelligent transportation field. The Paper gives a deep research on the development and status of the Plate license recognition system, analyzing the characteristics of the license plate and the difficulty of license plate location. On this basis, this paper proposes a novel and efficient algorithm of License plate location base on machine learning and edge-color pair. Firstly, using the technology of image preprocessing to deal with image of vehicle, including gray-scale transformation, noise and enhancing contrast, edge detection and so on, improving the visual effects of image, laying the foundation of License plate location; Secondly, detecting the General area of License plate base on edge texture, then extract Haar-like feature of this area’s gray image, using the Adaboost algorithm to locate the Plate license; If there is a number of Plate license’s candidate regions were located, detecting the edge-color pair of candidate regions base on characteristics of License plate’s color,then combining the texture feature of characters to sift License plate from a number of candidate regions.The performance of the method was investigated on many car images under various conditions. The result of experiment showing that the location method has higher accuracy, speed and robustness. The method has a certain application value else.