Detecting And Analysis Of Suspected Operating Vehicles Based On Electronic Registration Identification Data Of Motor Vehicle

Nowadays,the urban passenger transport system has been greatly developed,but it still can not meet the diversified travel needs of urban residents.In such an environment,suspected operating vehicles have emerged in the urban private car group.Suspected operating vehicle refers to the private vehicle that carries passengers for profit without permission,which greatly affects the order of traffic market and social harmony and stability.Therefore,it is very important to find out the suspected operating vehicles from the private car group and control them.At present,the means of checking suspected operating vehicles are mainly manual,which not only wastes a lot of human resources,but also is inefficient.Electronic Registration Identification of the motor vehicle(ERI)is a new traffic information collection technology,which has high collection accuracy and can cover all kinds of vehicles in the city,including private cars.Using the collected trajectory data to detect suspected operating vehicles can improve the detection efficiency and accuracy.Therefore,this thesis proposes an identification method of suspected operating vehicles based on ERI.On this basis,the hot routes of suspected operating vehicles is found,which provides a reference for relevant management departments to manage suspected operating vehicles.The main work of this thesis is summarized as follows:(1)This thesis analyzes the vehicle trajectory based on the ERI to prepare for the identification of suspected operating vehicles and the mining of their hot routes.First of all,the data quality of the ERI is preprocessed.On this basis,the trajectory data based on the ERI is defined and extracted.Finally,according to the time threshold,the partition method of the vehicle trajectory based on the ERI is proposed.(2)To solve the problem of identification of suspected operating vehicles,this thesis proposes a model based on logical regression.Firstly,from the definition,taking the taxi as a reference,analyzing its behavior features,then defining and calculating the features that can distinguish the ordinary private car and the suspected operating vehicle,then using the features to train the identification model of suspected operating vehicle,finally using the trained model to find the suspected operating vehicle set from the private car group,on this basis,the model is verified and we find out the hot area of suspected operating vehicles.(3)On the basis of the above two works,this thesis uses the method of trajectory clustering to find the hot routes of suspected operating vehicles,which provides a basis for analyzing the travel characteristics of suspected operating vehicles.Specifically,first of all,considering the space-time features of vehicle trajectory data,according to the features of ERI trajectory data,a calculation method of trajectory similarity measurement is proposed.On this basis,based on DBSCAN clustering algorithm,a trajectory clustering algorithm based on spatiotemporal similarity measurement,SPATIO-DBSCAN,is proposed.Through the algorithm,the hot routes of suspected operating vehicles are found,and the distribution of hot routes in three periods of the whole day is analyzed.