Design And Implementation Of Massive GPS Point Data Visualization System

As an important part of human activity trajectory data,GPS data not only describes the temporal and spatial change properties of moving objects,but also contains rich semantics.Through the visualization platform,large-scale GPS data can be presented to researchers more intuitively,and rich interactive operations can help them reveal the temporal and spatial laws and even deeper semantics contained in the data.In a large-scale GPS data visualization platform,the design is usually based on the Web-side visualization model,and the data structure is optimized to achieve efficient storage of index data,and GPU rendering is used to accelerate large-scale rendering.The visualization platform designed in this thesis mainly focuses on the back-end computing power and the browser's field of view characteristics in the visualization window.The multi-resolution pyramid model is used to realize the organization and management of large-scale data,and the large-scale data is realized by optimizing the coordinate conversion algorithm and using GPU parallel computing.Fast drawing.The platform can effectively reduce the system memory footprint,ensure that large-scale GPS data of more than 10GB is loaded on-demand,multi-scale browsing,rapid positioning and drawing,and efficient rendering even when the view window changes,improving the visual interactive operation experience.The visualization platform designed in this thesis can solve the current large-scale GPS data visualization problems such as high-performance hardware requirements,too long rendering time or even unresponsiveness,and poor interactive operation performance,which lays a foundation for researching users' analysis and decision-making on large-scale GPS data Solid foundation.The primary works in this thesis are presented as follows:(1)After fully considering the simple deployment,easy distribution,timely maintenance of the current visualization system,and the company's recruitment requirements for developers,the design principle of B/S as the system architecture is determined;(2)WebGL-based rendering architecture realizes direct hardware-based rendering,improves the performance of Canvas-based rendering,makes the entire rendering smoother and optimized,and improves system processing performance;(3)Improve the data front-end and back-end loading architecture of GPS point data visualization,use the multi-resolution pyramid structure to load the visually drawn data,and use the data slicing technology to improve the data processing and rendering capabilities of the entire system;(4)The use of GPU-based coordinate transformation algorithms improves the inefficiency of drawing performance during front-end interaction when using CPU to process data,thereby reducing drawing delay and improving user experience.Based on the Web platform,the system preprocesses large-scale data on the server side,and transfers part of the client's computing pressure to the GPU.From the current practice of this thesis,this method is suitable for large-scale GPS point data visualization.After a successful attempt,you can continue to explore based on this idea.