基于RBF的散乱点曲面重构  

RBF-Based surface reconstruction from scattered data

作　　者：李乐庆[1] 康宝生[1] 

机构地区：[1]西北大学信息科学与技术学院,陕西西安710127

出　　处：《西北大学学报（自然科学版）》2011年第2期221-225,共5页Journal of Northwest University（Natural Science Edition）

基　　金：陕西省自然科学基金资助项目(FC06121)

摘　　要：目的利用径向基函数(RBF)神经网络对非线性函数的逼近,以及强大的抗噪、修复能力等特点,对散乱点曲面进行重构。方法在已有RBF神经网络模型的基础上提出了采用相关系数的方法自动确定网络隐含层核函数半径大小。结果给出算法的分析和实现步骤,并将该模型应用于带有高斯噪声的散乱点曲面重构。结论与利用传统方法确定的RBF模型相比,该RBF网络模型的整体逼近精度更高,修复和抗噪性能更强,且得到的曲面光顺性好。Aim To use the RBF neural network′s advantages of approximating no-linear function,powerful antinoise and the capability of repairing,and to reconstructe surface from an unorganized cloud of points.Methods The method of the correlation coefficient between points was used to self-adjust the radius of the keral function in the network.Results The algorithm′s scheme and analyses were given and the proposed method was applied to the free surface reconstruction from an unorganized cloud of points involving noise.Conclusion Compared with the RBF model established by the traditional methods,this neural network can not only approximate the surface with high precision but also has powerfull anti-noise and the capability of repairing,and so on.The smoothness of surface is good.

关 键 词：径向基函数 散乱数据 曲面重构 神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]