Research On Image Preprocessing And Matching Of Vehicle License Plate Recognition

China’s car ownership has a gradually rising trend along with the rapid development of economy, Now, how to build a perfect urban traffic monitoring system has been widely used in the Department of traffic supervision departments, and gradually developed into a smart transportation system has been focused on the goal. We know that the realization of intelligent traffic can not be separated from the protection of traffic regulation, and the key position of regulation, both in terms of traffic control, or in terms of pattern recognition can not be replaced. In recent years, people’s income levels have greatly improved, at the same time, the quality of life have higher pursuit, so this will undoubtedly promote the automobile market in great demand, the performance of the number of private cars is increased year by year the situation, the attendant is to the onset of more and more traffic.Whether it is the emergence of a large number of traffic accidents, or in violation of traffic rules of behavior generally increased, to our city traffic management has brought enormous pressure and challenges. This occurred clearly reflects the traffic supervision departments lack of supervision or regulation on both. However in fact, even regulatory supervision and how to strengthen, is also very difficult to break through the constraints of the actual situation, is basically impossible for every city in every road and the junction of the realization of 24 hours without loopholes in a full range of monitoring. Therefore, if you want to get better overcome the problem need to take the high-tech means, by way of mutual integration of machine vision and image processing. The design and implementation of a suitable for urban traffic management of intelligent traffic monitoring and management system, complete the road motor vehicle license plate, mark, types of accurate monitoring and correctly identified.Now relatively mature motor vehicle identification methods have come out in succession, like some developed countries of the city’s many toll stations, had to get rid of the driver through the toll station must stop fee vehicles way, never trouble and special delay time, causing a certain degree of traffic congestion. At present adopts the technical means is the use of onboard equipment and ground base station equipment both automatic identification and registration, so as to achieve the purpose of charges, is the so-called etc(Electronic Toll Collection). This way, eliminating the need for a toll station workers and drivers division Fu procedures, reduce the artificial way may bring misoperation, greatly improve the toll station work efficiency and reduce the traffic jam. There are a more general method for the detection of motor vehicle comes with parameters, replant method is based on the premise, the vehicle according to certain standards to complete the classification using the appropriate classification algorithm dependent like laser, red,surface acoustic wave(SAW) and so on and so on the means of identification, a number of feature information of the motor vehicle, is the professional field, it is often said of vehicle automatic classification technology. Its advantage is that it is more convenient to use, so the application of a lot of scenes, but also more common. Derived from the microwave and ultrasonic detection and for vehicle recognition rate is more ideal, but the fly in the ointment is the installation process is very cumbersome, the equipment is often affected by the surrounding environment, so the cost is relatively high, if the device received damage,often resulting in more serious traffic gridlock.The MATLAB license plate on the license plate image matching algorithm is studied based on. Firstly, introduces the license plate image matching research status at home and abroad, followed by the analysis and comparison of the license plate image matching theory,and with the license plate recognition as an example of license plate image preprocessing technology:(1) license plate location: transverse preliminary positioning, longitudinal preliminary positioning, gray, two values, transverse precise positioning, vertical precise positioning;(2) license plate to the noise, character segmentation. Finally, feature matching and recognize the characters of license plate based on gray correlation. Firstly extracted character features, under the precondition of satisfying the feature matching calculation segmentation of each character in the character and license plate character library gray correlation factor and selecting the gray mutual factor related maximum as the recognition result and recognize the license plate character.