Research On The Realization Of Automatic Vehicle License Plate Identification Dechnology

With the rapid development of our economy and the improvement of the living standards, the national consumption of vehicles has enjoyed an escalating trend, which incurred our traffic in an awful situation, and the traffic administrative department faces severe problems, such as the problem of how to achieve the rapid and intellectualized traffic management and so on, therefore, the research topics about the automatic identification of the vehicle license plate technology have come into being. Under the precondition of insuring the smooth running of the vehicle in the course of driving, the automatic vehicle license plate identification technology refers to the identification and output of the image acquisition and character recognition quickly and accurately. Based on technologies such as digital image processing, mode identification and the computer vision etc., the automatic vehicle license plate identification technology can be widely applied in the following areas, such as getting through the roadblock and the bridge, handling the violations by the electronic police as well as the supervision of roadway and so on, consequently, the research of the ALPR technology plays an important role in the practical application. The content of this paper includes the following several respects:Apart from the structure of the automatic vehicle license plate identification system, the current research status at home and abroad has also given in-depth analysis by the paper. Based on the above-mentioned information, this paper elaborated the difficult problems in automatic vehicle license plate identification. Then the second chapter presented the image preprocessing, the positioning of vehicle license plate, normalization as well as the sloping and correction of image, etc., and the third chapter expounded the methods of the plate edge acquisition, frequency division and threshold segmentation, etc., what’s more, the remove of the license plate frame as well as the segmentation and normalization of the single plate character have also been specified on the basis of above-mentioned methods. The fourth chapter elaborated on the comparison of complexity, efficiency and accuracy index of the neural network identification method and template matching recognition method, and thus this paper put forward the specific methods of the integrated optimizing application based on it. The fifth chapter presents the readers with the structure and basic function of the prototype system derived independently by the author, which will provide the experimental platform for the practical algorithm application. Finally, in order to re-position the direction of the future work and study, the thesis summed up the main work and innovative research results as well as the disadvantages of the prototype system.