基于多分辨率数据源镶嵌的大规模地形实时绘制算法  

Real-time rendering of massive terrains based on tessellated multi-resolution datasets

作　　者：郎兵[1,2] 方金云[1] 韩承德[1] 

机构地区：[1]中国科学院计算技术研究所空间信息处理技术实验室,北京100190 [2]中国科学院研究生院,北京100049

出　　处：《高技术通讯》2009年第8期833-838,共6页Chinese High Technology Letters

基　　金：973计划(2004C13318202);863计划(2001AAl35210;2002AAll4020)资助项目

摘　　要：针对不同分辨率的测量地形数据源,提出了镶嵌四叉树模型,用以实现对多分辨率数据源的镶嵌,以及与细节层次(IOD)算法的融合。设计了基于此模型的批LOD算法,给出了相关公式,并设计了一种从CPU到GPU的渐进LOD数据传输方法。分析了镶嵌四叉树的场景浏览模式,设计了基于此模型的双向两级外存数据调度算法,通过水平、高度两个方向的预测和两级数据预取实现数据页的动态更新,实现了三维地形场景的快速构建和实时浏览。实验表明,该算法能够快速处理大规模多分辨率数据源并且能够充分发挥CPU、GPU与I/O的效率,从而在有限环境下进行快速实时绘制。For handling multi-resolution terrain data sources, the paper proposes a tessellated quadtree model to tessellate mul- ti-resolution terrain datasets and integrate level of detail （LOD） algorithms. An aggregated LOD algorithm based on the model is given for progressive transmission of vertices from CPU to GPU. By analyzing the roaming pattern of the tessellat- ed quadtree, a new out-of-core data paging and prediction algorithm is presented. The real-time loading of datasets is achieved by employment of two orientations of prediction and the double level paging strategy. The experimental results show that the proposed algorithm can handle large-scale multi-resolution data sources and balance the tradeoff between CPU, GPU and I/O, and realize real-time rendering under limited resources.

关 键 词：镶嵌四叉树(TQT) 细节层次(LOD) 渐进LOD传输 地形绘制 外存模型 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] P208[自动化与计算机技术—计算机科学与技术]