Spatio-temporal Visualization And Visual Analysis Based On Taxi Data

In recent years,with the advancement of the internet and internet of things technologies and the growing of urban population,urban data is exploding.Taxi GPS data is a kind of the trajectory data.The trajectory data contains people's daily travel habits,such as time information,regional information,the preference of taxi driver's route selection,speed and other travel informations.Data visualization technology is a technology that has gradually prospered in the era of big data.User takes use of the visual analysis to mine value of data which is presented in the form of graphical visualization.As trajectory data has spatiotemporal properties,visual analysis makes it easier to mine the value of trajectory data.At present,many scholars at home and abroad have studied taxi GPS data,mainly through the construction of models to study traffic behavior analysis,traffic state estimation and travel Origin-Destination(OD)prediction.However,study involved less finding out the pattern of tourist which is a kind of specific peple.through visual analysis.Moreover,taxi is the first vehicle for tourist in Hangzhou which was a famous tourism city.Therefore,the travel information of tourist has great value for city managers.Taking the West lake as our target area,we design a visual analysis system based on the taxi data from 2014.11.01 to 2014.11.07.Based on the system,we firstly analyzed the travel time pattern of the tourist in the scenic spot,and the travel area pattern is further analyzed through the heat map and the scatter plot guide.In OD part,the user can select the tourists by themselves and view the OD travel process of taxi dynamically.Then,combined with the trajectory distribution,time bar graph,and Point Of Interest(POI)firework graph,the city manager can obtain the travel route selection,travel time distribution,destination or source function area information of the visitor in the interactive exploration.In this thesis,we firstly introduce some related works at home,then we introduce the process of data cleaning and storage.Later,we introduce the principle of design in detail on visual components of our system.We elaborate on the factors considered of the visualization components,such as the granularity of data,time and space simultaneously visualization,interactive analisis of space and time,visual aesthetics,etc.A lot of pictures were displayed,which meet our study need of the travel time pattern,the travel area pattern,and the dynamic travel process of OD.Finally,we present two visual analysis cases in detail through a lot of pictures,then based on these pictures,we can find out travel patterns of tourists from some aspects,such as travel time pattern,travel area pattern,Point Of Interest(POI)information.In the last,we hope our visual system can be a usful tool for city managers.We can transform the data into available urban assets which can promote the construction of Hangzhou.