Research And Implementation Of Web-based Trajectory Data Visualization

With the rapid development of vehicle networking and satellite positioning technologies,different types of sensors continuously generate a large amount of vehicle trajectory data.The vehicle trajectory data contains important information such as speed,direction and position,and it also has time-space characteristics.Vehicle trajectory data has its important value in the fields of urban traffic planning,intelligent transportation,vehicle monitoring,personalized recommendation and other applications.This paper uses the trajectory data of Shenzhen private cars then adopts a plug-in-free,cross-platform visualization method based on B/S architecture,and develops a web-based trajectory data visualization system,which can be used to understand the space-time feature of trajectory data in a very intuitive way.Attributes are expressed over time,which facilitates the display of the dynamic evolution process of urban traffic hotspots and travel rules of vehicles,and provides the necessary foundation for urban intelligent traffic management and big data analysis and research based on vehicle trajectories.The main work of this article includes:Firstly,First deal with the original trajectory data,there are interference and noise problems.An abnormal data filtering algorithm is proposed to remove the invalid trajectory data such as cross-border and anomaly.For the missing trajectory data,a missing trajectory recognition algorithm based on distance and time thresholds is first proposed to identify the missing trajectory segment,and then the missing trajectory is complemented by means of mean filling algorithm to maintain the integrity and consistency of the data.Secondly,for the trajectory data clustering problem,Formation the thermal map and the starting point aggregation,this paper uses a simple and effective algorithm-core density estimation algorithm to solve problem,using the insertion method and the interleaving method to select the appropriate bandwidth parameters,experimentally proved that the clustering result of bandwidth parameters chose by the insertion method is the best.Then,In order to improve the development efficiency,the Django framework of the MVC pattern is adopted..The Django framework decomposes the view into views and templates that determine what data is presented and how it is presented.Instead of handling user input,you can simply call control-level functions to increase development efficiency.Finally,a web-based trajectory data visualization system is designed.and implemented.The system consists of four func.tional modules,which are the starting point,the heat map,the trajectory query,and the travel data statistics.The starting point is the statistic of the vehicle ignition and extinguishment points;the thermal map is the color density of the active vehicle which is represented by the color patches of different colors;the trajectory query provides the historical trajectory query function of the vehicle and the trajectory is reproduced;the travel statistics are the statistics of the number of trips of the active vehicle every day.After system testing,the desired design goals are achieved.