Research Of Group Travel Regularity Analysis And Vehicle Tracing Based On VLPR Error Detection And Correction

With the development of social economy in our country,more and more large-scale activities are being held in China.The problem of traffic congestion is becoming more and more serious.It is one of the problems that restricts economic development of China.ITS(Intelligent Transportation System)has become a worldwide trend.VLPR is one of the most important technologies in intelligent transportation,which plays an important role in traffic management,public security prevention and control,and the tracking and positioning of suspected vehicles.However,the accuracy of license plate recognition is affected by many factors.There are a lot of wrong license plate numbers in the license plate recognition data base and these huge amounts of data are not properly utilized.Based on Hangzhou city license plate recognition data,this paper presents a license plate number of error detection and correction algorithm.Through the analysis of a large amount of data,temporal and spatial variation of the typical city road traffic flow is studied.Based on the revised data,an optimized method for obtaining typical routes is studied.Besides,based on the license plate recognition data,application of Markov model in vehicle trajectory prediction is studied.The main work of this paper is as follows:(1)Based on vehicle recognition data and vehicle registration information data of Hangzhou city,the dynamic change law of traffic flow of typical road is studied,including the total flow,the local car traffic flow,the traffic flow of small cars and large cars,etc.This paper proposes a method of getting average travel speed and study the dynamic change law of road average travel speed.The results showed that the traffic flow of typical road has certain regularity in time and space.It can provide data support for the prediction of traffic flow and the formulation of traffic guidance strategy.Besides it has certain theoretical guidance to ensure the light traffic;(2)The recognition accuracy of the VLPR equipment is calculated based on the vehicle recognition data and analysis the influencing factors that affect the accuracy of VLPR.Besides this paper presents a license plate number of error detection and correction algorithm.This algorithm is used on the vehicle recognition data and verified by the road video data.The result shows that this method can detect the wrong license plate and has high success rate of correction,thus improving the accuracy of recognition equipment;(3)By analyzing the vehicle recognition data of a certain area of Hangzhou,this paper studies the threshold of the dynamic hot spot and generates the regional hot spot intersection.Based on this,a typical route algorithm is obtained.Then it determines the typical travel routes at different times in the area,and analyzes the law of citizens’ mass travel.This paper studies the application of Markov model in vehicle path tracking and prediction and gets the driving route of the vehicle under the GIS platform.Besides,it carries on the test to the established Markov model in vehicle trajectory prediction.The results show that the algorithm can obtain the typical driving route of different time periods,which has important practical significance to analyze the mass travel law.The model has a high accuracy rate of vehicle path prediction.