在三维空间直接进行的三角剖分算法  

A direct triangulation algorithm in three-dimensional space

作　　者：康子阳[1] 苏显渝[1] 刘天成[1] 

机构地区：[1]四川大学光电科学技术系,成都610064

出　　处：《四川大学学报（自然科学版）》2010年第3期505-510,共6页Journal of Sichuan University(Natural Science Edition)

基　　金：国家自然科学基金(60527001)

摘　　要：光学三维传感为各种应用提供了原始点云数据,对于大多数应用,例如逆向工程、CAD/CAM系统或计算机图形软件来说,把这些点转化成三角网格或参数描述是非常必要的.基于逐点插入法,生长法提出了一个融两种算法优点于一体,兼顾空间性与时间性的合成算法,并将其推广到三维空间.该算法直接在三维空间内实现三角剖分,省去了网络拼接过程.经测试,该算法运算速度快于逐点插入法和生长法.并且有效避免了"空洞"和"重叠"现象.Three-dimensional optical sensing provides original points cloud data for many applications. And for most of them, such as reverse engineering, CAD / CAM system or computer graphics software, it ~s necessary to change these points into the triangular grid, or the parameters description. Base on incremental insertion, triangulation growth, this article provides a new upgrade algorithm--compound algorithm. The new algorithm takes advantages of them. It uses computer resources of time and space more reasonably. As well as extending to the three-dimensional space. The algorithm achieves 3D reconstruction directly in the three-dimensional space, without network connection. Through test, its running speed proves faster than that of incremental insertion and triangulation growth. And this new algorithm can avoid the ＂hole＂ and ＂overlap＂ phenomenon effectively.

关 键 词：DELAUNAY三角化 逐点插入法 螺旋边生长法 合成算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]