基于外存八叉树的三维激光点云实时渲染技术  

Real-time rendering of three-dimensional laser scanning point cloud based on out-of-core octree

作　　者：邱炎[1] 梁玉斌[1] 崔铁军[1] 

机构地区：[1]天津师范大学城市与环境科学学院,天津300387

出　　处：《天津师范大学学报（自然科学版）》2017年第3期58-61,共4页Journal of Tianjin Normal University：Natural Science Edition

基　　金：天津师范大学博士基金资助项目(52XB1306);天津师范大学市级工程中心开放研究基金资助项目(YF11700103);天津师范大学应用开发研究基金资助项目(52XK1002)

摘　　要：为了解决海量点云数据的实时渲染问题,采用基于外存储器构建细节层次八叉树索引结构的方法用于三维点云数据实时渲染,并用8G内存的普通计算机对天津师范大学教学区10.2GB的扫描点云数据进行实验验证.结果表明:实时渲染时的帧率在16 f/s上下浮动,且无明显延迟,表明该方法支持超过内存容量的三维激光扫描点云数据的实时渲染.In order to solve the problem of real-time rendering of massive cloud data, the method which is based on octree index structure in different levels of detail established on the external memory is used for real-time rendering of 3D point cloud data. Experiments on scanning point cloud data of 10.2 GB in teaching area of Tianjin Normal University were done with ordinary computer with 8G memory. The result shows that the frame rate of real-time rendering fluctuates at 16 f/s, and there is no obvious delay. Experiment results show that the presented method supports real-time rendering of three-di- mensional laser scanning point cloud data whose volume is larger than the capacity of the physical memory.

关 键 词：实时渲染 八叉树 层次细节模型 三维激光点云数据 

分 类 号：P237[天文地球—摄影测量与遥感]