摘要
文章使用成长型神经网络来实现曲面的快速重建,在重建过程中采用线性组合的方式分裂节点;使用一种简化的方法计算节点的Voronoi区域面积,以用来分配新增节点的计数器值;最后对生成的曲面进一步综合优化。实验结果表明,使用该算法能够快速获得形状良好的曲面。
In the paper, surface reconstruction is realized by using Growing Cell Structures(GCS). During the surface reconstruction, the node is split using the method of linear complexity,and a simple approximate method is used to calculate the area of Voronoi cells so as to assign a signal counter to the new node. The surface is finally optimized by a compound method. Some experiment results are given, which show that the method presented in this paper is effective and rapid in surface reconstruction.
出处
《合肥工业大学学报(自然科学版)》
CAS
CSCD
北大核心
2006年第8期984-987,共4页
Journal of Hefei University of Technology:Natural Science
关键词
神经网络
自组织映射
成长型神经网络
neural network
self-organizing mapping(SOM)
Growing Cell Structures(GCS)
作者简介
王世东(1975-),男,安徽合肥人,合肥工业大学硕士生;
张佑生(1941-),男,湖南浏阳人,合肥工业大学教授。博士生导师