点云模型的切片式曲面重构算法设计  

DESIGN OF SLICED SURFACE RECONSTRUCTION ALGORITHM FOR POINT CLOUD MODEL

作　　者：陈华伟 袁小翠[2] 伍权[1] Chen Huawei;Yuan Xiaocui;Wu Quan(School of Mechanical and Electrical Engineering,Guizhou Normal University,Guiyang 550025,Guizhou,China;Jiangxi Province Key Laboratory of Precision Drive and Control,Nanchang Institute of Technology,Nanchang 330099,Jiangxi,China)

机构地区：[1]贵州师范大学机电工程学院,贵州贵阳550025 [2]南昌工程学院江西省精密驱动与控制重点实验室,江西南昌330099

出　　处：《计算机应用与软件》2020年第8期245-249,288,共6页Computer Applications and Software

基　　金：国家自然科学基金项目(51365037);江西省教育厅项目(GJJ14128)。

摘　　要：对大数据量、复杂点云模型进行曲面重构,切片法具有速度快、算法简单的优势。采用等区间分割和半区域点云映射法获得分割面上的平面点云,并采用转角法提取切片点云凸包边界;提出基于边界闭合性分析的闭合边界解环方法,以及边界起点和方向一致性调整方法;基于拟合曲线及采样型值点进行NURBS曲面拟合。通过程序开发实现上述算法,模型实验和对比实验表明,顺应模型主轴方向的分割有助于更准确提取切片点云边界,切片密度越大则拟合曲面的精度越高。To reconstruct surface for big data and complex point cloud model,slicing method has the advantages of fast speed and simplicity.Planar point cloud on the segmentation surface was obtained through the bounding box and half zone point mapping,and the convex boundary of sliced point cloud was extracted via planar vector angle method.We proposed a closed boundary loop method based on boundary closure analysis and a method for adjusting the starting point and direction consistency of the boundary.Then,NURBS surface fitting was based on fitting curve and sampling point.Our algorithm was realized via program development.Model experiment and comparative experiment manifest that slicing along the main axis of the model can help accurately extract point cloud boundary,and higher density slices can gain more accurate fitting surface.

关 键 词：逆向建模 NURBS曲面 切片法 边界提取 

分 类 号：TP391.7[自动化与计算机技术—计算机应用技术]