License Plate Recognition System Key Algorithm

As coming along with the development of the highway traffic enterprise in the new century,the requirement of the Intelligent Traffic System(ITS) gets into a new level.The vehicle license plate is the clearest,exactest,and exclusive symbol.License plate recognition technology which is a special vision system,can automatically record the dynamic data of the Vehicle traveling,judge and extract the image data of vehicle license plate effective,and also can recognize the character of vehicle license plate precisely. Thus,Vehicle License Recognition system play a vital role in ITS.A Vehicle License Recognition system usually,comprises three parts,vehicle license locating,character segmentation and character recognition.In the license plate recognition system,these phenomenon of blur,incline and illumination odds because of weather or shooting angle acquired great hardness.Point to these issue,an adaptive method for preprocessing is described,which have the characteristics of fast,good result and simple.It is suitable for the license plate characters preprocessing and it also facilitate the subsequent character segmentation and character recognition.Vehicle license locating and character segmentation play the crucial role,which influence the precision of the whole system directly.Thus,this thesis studies the vehicle license localization and division based on above vehicle license preprocessing.In this thesis,two vehicle licenses preprocessing methods are studied.The first one is the gray vehicle licenses preprocessing.A new preprocessing techniques based on histogram equalization of wavelet package division is proposed.The experiment results showed that this algorithm achieved the effects of edge enhancement,noise cancellation as well as time complexity reduction.The traditional image enhancement algorithms usually convert to gray image firstly and at the same time,these algorithms enhance noise signal in image while they enhance image.With the combination of the characteristics of wavelet package and the predominance of histogram equalization,a novel method of color license plate enhancement method is presented here.The experimental results showed that this algorithm enhanced license plate images of high brightness effectively and the time cost was greatly reduced.This thesis studies the image segmentation based on edge detection and threshold. Based on the above segmentation methods,vehicle license location and division is studied.The location of the characters of vehicle licenses is found with our method based on adaptive threshold method and projective transformation.