Research And Application Of Real-Time Rendering Of Complex Network Data Based On 3D Render Engine

The traditional way to visualize network data is text and form,and a small amount of points and lines to present network structure;it is unable to meet the increasing expansion network data analysis and management,as an effective tool to help users to understand the network structure,and mining the hidden information in the network data.In many fields,it can be analyzed with complex network visualization techniques.As an important information visualization technology,complex network visualization technology makes the best of human visual perception system to display data in graphic way,help us quickly and intuitively to interpret and browser a complex network structure.Frist of all this thesis introduces the research status of complex network visualization and summarize the existing complex network visualization solutions.It implemented existing visualization algorithms and increased the visual dimensions of complex networks and implemented a 3D complex network rendering engine and applied to the actual complex network could render the complex network in real time.The main work is as follows:1.The adaptive layout of complex network graph: in large-scale complex networks,based on social network graph,we study and implement the maximum entropy model based on force directed layout in three-dimensional space to visiualize the relationship between user groups2.Research on the improve readability of complex network graph: After layout a complex network,there are lots of edges connect large scale points,edges are intersecting and covering each other.In this thesis,we study and implement multi-level edge binding algorithm in three-dimensional space,optimize the layout effect of complex network,and reduce the intersection and coverage of edges.3.Visualization of spatiotemporal data.Based on the historical data of vehicle driving in urban traffic,the GPS data of vehicles are visualized in real time in the 3D rendering engine implemented in this thesis to simulate the driving condition of vehicles.4.Implement the 3D network rendering engine: Based on OpenGL API,implement a 3D network rendering engine,which integrates the maximum entropy layout algorithm and multi-level edge binding algorithm in 3D space,and use GPU to render complex network in real time.The three-dimension network rendering engine implement in this thesis is applied to large scale complex networks.such as Facebook social networking data,Wiki-Vote voting data and so on.The advantages of network visualization in three-dimension space are analyzed.And use GPS data to simulation running vehicle shows the ability to render in real time.