Car License Plate Automatic Identification Technology Research

Car license plate recognition (License Plate Recognition, LPR) technology is a comprehensive application of computer technology and digital image processing technology in the intelligent traffic management system, is a key link in the intelligent traffic management. License plate recognition system contains the license plate location, character segmentation and character recognition of three parts, the paper these three main aspects of research, analysis of the advantages and disadvantages of the common algorithms, combined with the license plate of their own characteristics, and finally to design an effective algorithm to complete license plate recognition.1.License Plate Location Research analysis on the the Common license plate positioning method, using a lattice of a certain range of colors in the image of a license plate positioning method of color component that is based on statistics collected the outline of the license plate area.2. Character Segmentation Character segmentation method used in this article, the first image from left to right scan, the statistics of the percentage of the number of black spots appear in each column with respect to the image height; then from the bottom up scan line where the scanning lines for the third dimensionprojection can be drawn from a number of consecutive black block, you can believe that the license plate number of black blocks of seven horizontal split; last character of a transverse split vertically information extracted positioning the left and right borders of the known charactersextracted the vertical boundary information.3. Character Recognition The character recognition is a key technology in the system, and the need to complete the license plate letters, numbers, and character recognition, analysis of the characteristics of license plate character, a combination of approaches to use two algorithms to complete the identification. Using character recognition method based on the minimum distance classifier; license plate character recognition approach based on artificial neural networks for Chinese characters, letters and numbers. To improve the efficiency of the recognition by the combined use of these two methods, but also improve the recognition accuracy.Through experiments that the algorithm used in this article, which can effectively and accurately complete license plate recognition.