平行散点图：基于GPU的可视化分析方法  

Parallel Scatterplots:Visual Analysis with GPU

作　　者：杨珂[1] 罗琼[2] 石教英[1] 

机构地区：[1]浙江大学CAD&CG国家重点实验室,杭州310058 [2]香港科技大学计算机系,中国香港

出　　处：《计算机辅助设计与图形学学报》2008年第9期1219-1228,共10页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家“九七三”重点基础研究发展规划项目(2002CB312105)

摘　　要：提出一种分析多维数据集之间关系的信息可视化方法——平行散点图.结合平行坐标、散点图方法,综合了焦点+背景、多视角、多视图、刷子等交互与可视化策略,使人有效地观察与分析多维数据集之间的连接关系;利用统一渲染GPU对粒子、线段、公告牌等的绘制与输出功能,在千万级数据集上达到较强的深度感与交互级的绘制性能;提出GPU上基于空间填充曲线的聚类算法,可交互式地降低连接线的视觉杂乱度;将连接、聚类与可视化整合为一个基于GPU的系统,在千万级数据集上达到交互级的可视化分析.We propose an information visualization method, the parallel scatterplots, to analyze the relationships among multidimensional datasets. By combining the methods of parallel coordinates and scatterplots, the relationship join analysis among muhi-demensional data sets is effectively carried out via a variety of techniques such as focus＋context, multi-perspective multi-view, brush etc. Using the unified shader GPU to render and output particles, line segments and billboards, we have achieved strong depth cue and interactive performance on rendering tens of millions of data. To reduce the visual cluttering on large data, we propose a highly efficient clustering algorithm based on space-filling curves on the GPU. We have integrated joining, clustering and visualization on a GPU-based system which is capable of performing interactive visual analysis on tens of millions of data.

关 键 词：信息可视化 GPU 多维数据 连接 聚类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]