The Research About License Plate Recognition

With the rapid envelopment of domestic economy, number of cars, buses, trucks and other vehicles increases a great deal. More and more funds are put into the construction of roads, highways, parks by our government, which are essential to keep the booming market. Roads become broader and parking area have more space for drivers. However, as a very important part ofTransportation system, management dose not follow the steps of hardware constructions. As in the old days, almost everything is in the charge of human beings. To utilize the convenience of modern transportation, a way should be found to make good use of the existing transportation constructions.Intelligent Transportation System (ITS), such as Automatic Traffic SignalControl System, Global Position System (GPS), Navigation System, Intelligent Residential Area Management System , plays a increasingly important role in the management of transportation. Because license plate is efficient in identifying vehicles, License Plate Recognition (LPR) is fundamental in ITS. In recent years, as the technology of manufactory progress, electronic equipment gets higher performance. More theory in Digital Image Processing and Pattern Recognitions is put into practice .With the help of this progress;the management in modern transportation system gets all the essential conditions to make another giant progress.In this paper, I analysis the status of present transportation system, discussed some details in the field of License Plate Recognition .After all designed out a simulator system and realized the key modules of this system. Additionally ,the prototype includes two parts .They are Image Collection Module(ICM) and Image Processmg Module(IPM ).ICAO is mainly consisted of hardware such as cameras, switches, video card and cables. It monitors the video signal and cuts frame pictures at appropriate time . IPM, which is the key part of the prototype ,is realized by software .To realize it in a simpler way, it is coded in the environment of Mat lab.My work is focused on the Image Processing Module, where license locating and character recognition are realized. To locate the license plate, I took two steps,they are pretreatment and locating. The first is done to pop the plate-area out .In the second step, according to the frequency characters of the plate, I realized the locating in level direction in the way of scanning bean;and finished the vertical direction by using both the vertical projective image and frequency characters of the plate. I took the way of 13 characters in the process of character-acquiring.According to the features of license plate, the characters in the plate are divided into two groups, Chinese characters and none Chinese characters. ANN is the key tool in recognition for none Chinese charactersThere are eight chapters in this paper. The background of this paper will be introduced in Chapter One. In Chapter Two, the modules and the main flow chart are listed. Chapter Three some details are given to pretreatment of the whole process. Chapter Four is focused on the location method of license plate;while Chapter Five is about the segmentation of the characters. And Chapter Six is a detail introduction of how to pick up the characters form the letters. The Chapter Seven are details of recognition of none Chinese characters. Chapter Eight gives out the result of this experiment and also an evaluation of our system;disclose some weakness and point out some general solutions .The last chapter is a summary of my work.This paper will refer to the fields of Digital Image Processing, Pattern Recognition and Artificial Intelligence, etc.