一种基于神经元增长结构(GCS)的空间散乱点三角剖分方法  

A New Triangulation Method of Scattered Data Based on GCS

作　　者：张力宁[1] 张定华[1] 

机构地区：[1]西北工业大学现代设计与集成制造技术实验室,西安710072

出　　处：《机械科学与技术》2005年第7期820-823,864,共5页Mechanical Science and Technology for Aerospace Engineering

基　　金：航空科学基金项目(00H53076);国家教委博士基金项目(20006992)资助

摘　　要：空间散乱数据点的三角剖分是逆向工程的关键技术,本文提出了基于一种动态神经网络———神经元增长结构(GrowingCellStructure)的空间三角剖分新方法。这种方法具有可以直接处理带有噪声的数据,无需对数据进行平滑处理,网格规模可控,算法简单,生成的三角网格形状好等优点。Triangulation of scattered data is a very important problem in reverse engineering research. Based on a kind of dynamic neural network——Growing Cell Structure(GCS), a new method of triangulation is presented in this paper, and some modification processes are also given in addition to constructing a triangle mesh with good shape. Compared with other triangulation methods, this method is relatively easy to realize and can dispose of noisy 3D data directly. The amount of vertexes in mesh can also be calculated with this method, so we can control the size of mesh in advance. Finally, a practical example is given to prove its effectiveness.

关 键 词：网格增长结构(GCS) 三角剖分 神经网络 逆向工程 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]