| While the process of urbanization gives people the modern life,it brings many problems and challenges,such as the shortage of resources,the increase of energy consumption,traffic congestion,and ecological deterioration,etc.The traffic problem has always been one of the main problem affecting and restricting the development of the city.With the continuous growth of the number of motor vehicles,cities all over the country are facing the problem of urban road congestion in different degrees.The traffic hot spots can effectively reflect the traffic congestion in the city,which is of great value for the decision making in the aspects of urban planning,traffic management,and user investigation.In recent years,with the rapid development of wireless communication and data mining technology,the data acquisition and information processing ability of the path constrained by road network is increasing.As a new branch of data mining,trajectory mining has attracted more and more scholars' attention and research.In the study of trajectory mining,the use of trajectory clustering to analyze traffic hot spots is one of the most important applications in the study of mobile trajectory,which is mainly used to solve the problem of urban traffic congestion.At present,the focus of the research on trajectory data is the change of the spatial position of the moving objects,the time characteristics of the trajectory data are often involved in the processing of the trajectory data as auxiliary decision information,and the results of the trajectory clustering are not sufficiently applied.Therefore,there are still some problems in the path clustering and the traffic hot spot analysis.Research,analysis and optimization.In this paper,the spatial and temporal characteristics of the moving objects under the road network constraints are considered,and the cluster analysis method is used to excavate the hot paths and regions of the trajectory.The main research contents include the following four aspects:(1)Improved trajectory division method: according to the existing path division method,the standard of trajectory division is improved.In the path planning process,extracting feature points of Road intersection measurement in addition to information,increase the stay point of information,so as to further improve the efficiency of path division,the subsequent experiments show that the improved method can track division effectively reduce the number of trajectory data points,greatly reduces the storage space trajectory data.(2)Trajectory similarity measurement based on spatio temporal information: because most of the current researches are based on spatial information trajectory clustering method,a trajectory similarity measurement method based on spatio temporal information is proposed in this paper.Experiments show that the trajectory similarity measurement based on spatio temporal information can effectively maintain the space and time semantics of moving objects,making the clustering process more intuitive.(3)The traffic hot path and region extraction algorithm: In this paper,a traffic hot spot extraction algorithm is proposed based on the spatio-temporal similarity measure of LCSS to cluster the trajectory,proposes a hot path extraction algorithm based on the weight analysis of the sub trajectory.On the other hand,the hot spot is extracted by the clustering algorithm based on density,thus the analysis of the hot spot is realized.(4)The hot route and regional analysis of taxi trajectory in Chengdu area: the taxi trajectory data not only records the driving condition of the vehicle,but also reflects people's daily travel behavior.Through the cluster analysis of the taxi trajectory data of Chengdu,the hot areas and paths of residents' daily travel will be found,analyzes the hot spots and routes of residents' trips in Chengdu on weekends and workdays,as well as the early,middle and late peak.which can be used to solve the problems of traffic congestion in Chengdu.The auxiliary decision information will be provided... |
PDF Full Text Request