一种基于成长型神经网络的曲面重建快速算法  

A rapid algorithm for surface reconstruction based on Ggrowing Cell Structures

作　　者：王世东[1] 张佑生[1] 偶春生[1] 谢颖[1] 

机构地区：[1]合肥工业大学计算机与信息学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2006年第8期984-987,共4页Journal of Hefei University of Technology：Natural Science

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.

关 键 词：神经网络 自组织映射 成长型神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]