The Application Study Of Parallel Coordinates Edge Bundling Technology In Visualization

This is an era of data,data generated from human production and life is growing at an unprecedented pace.Visualization analysis technology which combined the strong cognitive abilities of human for graphic image and powerful data processing power of computer.It takes full advantage of human and computer and can help people efficiently and rapidly mine information hidden in data.In numerous visualization methods of multi-dimensional data,parallel coordinates is one of them,which can map multi-dimensional data to 2-dimensional space and each data is mapped to an edge.However,when scale of data set is very large,the edges of parallel coordinate will crisscross and overlap together with each other which will cause visual clutter.Edge bundling algorithm offers an important idea for solving visual clutter of parallel coordinate.However,the existing edge bundling algorithm can’t accurately express the amount of data information and may miss visual continuity among dimensions.In order to solve the above problems,two research contents were included in the thesis:1.This thesis propose an edge bundling algorithm to solve the problem.Every cluster in every axis is further divided to several groups in terms of different connection.Then,the connection between groups is built on reordered groups.Finally,all groups are rendered independently with different colors.In order to enhance visual effects,we des ign an interaction to highlight the selected group.Experiments show that this thesis approach is effective to visualize large data set and visual continuity of connection relation is pr eserved well.2.Design and implementation a multi-dimensional data visualization system based on this algorithm.This system supports two data import methods that access internet data and local data,the user interface can visually control parameters then visualize and analyze multi-dimensional data conveniently by parallel coordinates technology.In this system,the multi-dimensional data can be visualized by traditional parallel coordinates and edge bundled.At last,all research work of finishing the thesis were concluded,direction and target of further research were put forward.