散乱点云特征面拟合与求交算法  

ALGORITHMS OF FEATURE SURFACE FITTING AND INTERSECTION FOR SCATTERED POINT CLOUD

作　　者：陈华伟[1] 袁小翠[2] 吴禄慎[1] 

机构地区：[1]南昌大学机电工程学院,江西南昌330031 [2]南昌工程学院机械与电气学院,江西南昌330099

出　　处：《计算机应用与软件》2017年第12期245-249,共5页Computer Applications and Software

基　　金：国家自然科学基金项目(51365037)

摘　　要：逆向建模的主要目标就是通过曲面重构,向CAD输入NURBS等曲面模型。曲率是曲面的基本信息,采用二次曲面法估算点云曲率,结合曲率法和统计法对点云进行特征型面分割,有效识别了平面、圆柱面和球面等规则曲面。采用最小二乘拟合法求解曲面参数,拟合NURBS曲面,并采用Newton-Raphson迭代法求解面与面的相交线。实验中规则模型的特征面识别率达到100%,复杂规则几何模型的主要特征面能正确识别。实验结果表明该方法在以规则型面为主要特征的零件模型重构应用中的有效性。Surface reconstruction and its output of NURBS surface for CAD applications is the main object of reverse model. Curvature is the basic information of surface and the key parameter in surface reconstruction. In this paper, quadric surface method was used in curvature estimation for point cloud. By combining curvature method and statistical method, the feature surface of point cloud was segmented, and the regular surfaces such as plane, cyhnder surface and sphere were identified effectively. Moreover, the least square fitting method was used to solve the surface parameters, and the NURBS surface was fitted. The Newton-Raphson iterative method was used to solve the intersection llne of surface and surface. In the experiment, main features had been recognized and even 100% recognized for rule models. And the main feature planes of complex regular geometry model can be correctly recognized. The result manifests that the proposed methods are effective in surface reconstruction, especially for models composed mainly of rule feature surfaces.

关 键 词：曲率估算 曲面拟合 曲面求交 曲面重构 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]