Vehicle Fuzzy Pattern Recognition In Intelligent Transportation System

With the Economic development of our country,the demand of traffic enhanced continuously,while the traffic management is relatively lag,that makes the traffic jam increasingly serious,the way to improve it is introduce the intelligent transportation system.In this paper,we did some research of pattern recognition of vehicle automatic in the intelligent transportation system and presented the method of establish recognition model.There are many kinds of vehicle in our country and there no distinct difference among them,so it's very difficult to get a satisfied effect by using the conventional recognition method.In this paper,we establish a method to vehicle recognition using the two stages and character separation.The character of vehicle is so many and no distinct difference among these style,so in this paper we introduce fuzzy degree for character distill and using fuzzy degree as index of character for contribution of classification. First stage, we carry out the classification of truck and car by the support vector machine theory. Second stage, we respectively set up recognition model of car and truck by fuzzy neural networks,at the same time,we proposed using genetic algorithm optimize membership function and reasoning rule for solving a disadvantage that neural networks easily get into local minimum,that made systems has a good stability and accuracy.In this paper the character data was classified into train data and test data, the train data was used to trained and modified the model parameters,and the test data was used to testfied the effect of the recognition model. The result of experimentation testify the means has validity.The theory of this paper is propitious to accomplish automatic vehicle recognition,which lay a foundation for realizing electronic toll collection systems.