A Visualization-Oriented Trajectory Data Compression Method

With the rapid development of the visualization technology of trajectory data,making use of interactive display is of great significance to the visual analysis of trajectory data and to assist decision-making.However,the growth of global positioning systems and mobile terminal devices makes growing trajectory data have some characteristics,such as massive features,strong temporal and spatial sequence,and data redundancy,which has brought huge challenges to the visualization of large-scale trajectory data.However,large-scale trajectory data visualization is the primary prerequisite and performance bottleneck for large-scale trajectory data visualization analysis,which also makes large-scale visualization analysis face unprecedented pressure.Therefore,it is urgent to study scientific and efficient trajectory data compression methods in order to improve the visualization efficiency of large-scale trajectory data,which can better ensure the performance and visualization of trajectory data visualization,and to provide a reasonable basis and technical methods support for visual analysis of trajectory data.In order to better meet the needs of visualization analysis,this paper takes trajectory data as the research object and proposes a visualization-oriented trajectory data compression method.In view of the problem of the traditional trajectory data compression methods,which mainly consider the optimal compression rate instead to fully consider the shape retention of the trajectory after compression.Based on the visual analysis of the trajectory shape,this paper studies the trajectory data compression method that takes into account the visualization performance and visualization effects,which can provide efficient solutions to support large-scale visual analysis of trajectory data.The main research contents of the thesis include the following three aspects:(1)Trajectory data compression method based on Fréchet distance: Facing the need for fast visualization of large-scale trajectory data,this paper studies the visualization-driven trajectory data compression method taking into account the similarity between the compressed trajectory and the original trajectory,which think of the spatial characteristic relationship of the trajectory data as the main factor.The principle of this method is that uses the Fréchet distance as a distance measure and combines the trajectory Spatial-temporal scale characteristics.So on the basis of the principle,a trajectory data compression method based on Fréchet distance is proposed to realize visualization-driven trajectory data compression,which provides method support for the establishment of a LOD Model of Trajectory Data Constrained by Visual Domain.(2)LOD model of trajectory data constrained by visual domain: To solve the problems of low visualization efficiency and long response time of large-scale trajectory data,which may cause visual analysis performance bottlenecks,this paper combines LOD technology and fractal theory to research the calculation method about how the LOD of the trajectory data should establish the level.So a trajectory data LOD model with visible domain constraints is established which use the space to exchange the time to achieve efficient visualization of trajectory data and to solve the problem of low-efficiency visualization of large-scale trajectory data.The model provides technical support for the related visual analysis of the large-scale trajectories data.(3)Visualization experiment and analysis of trajectory compression: Using the real trajectory data of taxis in Chengdu as the experimental basic data to experiment,the trajectory data compression method proposed by this paper is compared with the existing classic compression methods.Analyze the comparison result and verify the validity and scientificity of the method in this paper.