摘要
该文提出了一种构造带权点集的Regular三角化的算法,此算法统一了构造点集的Delauany三角化的Bowyer/Waston算法.如果采用一种称为Delaunay树的数据结构来辅助点的定位,则算法的效率为 O(nlogn+n[d/2]).与Edelsbrunner和Shah提出的局部变换算法相比,此算法从理论和实现两方面都要简单一些.算法可以应用到曲线曲面重构和分子建模等领域.
This paper puts forward an algorithm for constructing the regular triangulation of a set of weighted points in Ed. The algorithm unifies and extends the Bowyer/Waston algorithm for the Delaunay triangulation construction. The time complexity of this algorithm is equal to O (nlogn+n[d/2]), if an auxiliary data structure called Delaunay Tree is used in it. Compared with the local transformation algorithm proposed by Edelsbrunner and Shah, this algorithm is simpler in theory and implementation. It can be applied to the fields of curve/surface reconstruction and molecule modeling.
出处
《计算机学报》
EI
CSCD
北大核心
2002年第11期1243-1249,共7页
Chinese Journal of Computers