Research And Application Of Licnese Plates Location Based On Morphological Mothod

In recent years, withthe high-speed development of our country’s economy and society, Chinese transportation industry will also usher in its own changes. As an important part of modern automated traffic management systems, intelligent transportation systems will also become more perfect. The automatic vehicle identification system which plays a very important role in intelligent transportation systems, is of realistic significance in maintaining the traffic safety and urban public security, preventing traffic congestion and achieving automated traffic management. It usually consists of image preprocessing, license plate locating, characters segmentation and character recognition. Among them, the license plate location is the key step, the latter license plate recognition can be smooth, and license plate recognition with high accuracy can be guaranteed, which is closely connected with the license plate location.In this paper, a full analysis of the various license plate location method proposed by others. And summarized the advantages and disadvantages of the various methods of license plate location. At the same time, combined the characteristics of the license plate with the difficulties of the existing license plate localization in china at present. In this paper, a secondary license plate localization way of the rough location first andprecise location later is proposed. whichuses mathematical morphology method and line scan method for coarse positioning plate and uses Hough transform method for precise positioning plate. The main works of this paper are as follows:1.Using mathematical morphology method and line scan methodto implement processing of coarse positioning plate:Firstly, after normalization and gray level transformation of the collected color plate image. Sobel operator is used to detect the edge of the car image, Secondly, the methods of erosion and closing operation in mathematical morphology are adopted to process the binary image, and several candidate license plate areas are obtained. Then, removing false license plate area by the line scanning method. Get reasonable area of the license plate location. Achieving coarse location of license plate.2. Using Hough transform method to implement processing of fine positioning plates: Firstly, median filtering and canny operator are used to process the license plate by coarse location. It can remove the noise and get a clear image contour. Then Hough transform is applied to locate the plate precisely. It can get an obvious rectangular border by Hough transform, then find the license plate up and down around the borderto achieve accurate location of license plates.In order to verify the feasibility and high efficiency of the proposed license plate locating method, a large number of simulation experiment are conducted in the MATLAB platformthrough the programming and design. Two hundred different types of images which are collected under different background are used in the experiment by comparing the results of the proposed method and other method. The experimental results show that the method can achieve the purpose of precise license plate location.