Research On Urban Dynamic Perception Technology Based On Multi-source Traffic Data

With the advancement of the urbanization process in the world,a large number of people are concentrated in large cities,and the rapid accumulation of population has triggered a series of urban problems,such as insufficient development space and unreasonable resource allocation.In order to improve the efficiency of urban management,official decision makers often need to perceive the regular movement of people in the city and the changes in urban dynamics caused by irregular movement of people.Urban planning and construction need to adapt to urban dynamics in order to promote urban development.Therefore,An understanding of the urban dynamics is crucial for official decision makers and urban planners.The origin-destination(OD) data sets generated by human daily travel behavior reflect urban dynamics.Previous spatio-temporal analysis methods utilize these data sets to extract popular city areas,with the ignorance of the flow relationships between areas.Several methods have been unable to determine time steps with similar spatial characteristics automatically,or failed to recognize the evolutionary patterns of various modalities for a city.In order to discover the hidden semantic-level urban dynamics from OD data,we propose a visual analysis method.The method carries out spatial simplification,and constructs a sequence of location networks at first.Then,the hourly network is studied as a document consisting of trip relationships among location clusters,enabling a semantic analysis of the OD data set as a document corpus.Hidden traffic topics are identified through a topic modeling technique in an unsupervised manner.Finally,an interactive visual analytics system is designed to intuitively demonstrate the probability-based thematic information and the evolutionary activity patterns of a city.People use different modes of transportation to reach different locations for different purposes,so it is not possible to accurately grasp the city dynamics from a single mobile data set.Besides,the spatial distributions of different types of Point-Of-Interest(POI) reflect the land use types,witch also indicates to a certain extent the purpose of people reaching a certain place.How to jointly analyze the multi-source mobility data sets and POI information is a great challenging.In this paper,we present a visual analysis method that explores urban dynamics based on multi-source mobile datasets and POI datasets.The method firstly constructs a region-feature-time tensor for each traffic dataset,and classifies the population movement pattern by tensor decomposition.Then,the visual analysis system is designed to progressively analyze urban dynamics from multiple angles,with POI-mobility glyphs visualizing multi-source datasets in a compact manner.Finally,case studies based on real data demonstrate the effectiveness of our approach.