基于移动最小二乘法的点云叶面三维重建  

Three-dimensional Reconstruction of Point Cloud Leaf Based on the Moving Least Square Method

作　　者：刘俊焱 云挺[1] 周宇[1] 薛联凤[1] 

机构地区：[1]南京林业大学信息科学技术学院,江苏南京210037

出　　处：《林业机械与木工设备》2014年第9期37-41,47,共6页Forestry Machinery & Woodworking Equipment

基　　金：国家自然科学基金"基于激光点云和视觉计算的阔叶树林学参数反演研究"(31300472);江苏省自然科学基金"激光点云数据中活立木林学生物参数解析研究"(BK2012418)

摘　　要：利用地面激光扫描仪获取户外树木的大量点云数据,从中截取树叶点云数据并以此进行曲面拟合,构建树叶真实的三维模型。主要针对空间散乱点云数据的曲面拟合方法进行研究,并利用Delaunay三角剖分构建树叶三维模型。在移动最小二乘法的二维曲面拟合方法基础上,针对空间散乱点云数据,提出了新的曲面拟合方法。通过移动最小二乘法对点云数据曲面拟合,得到了理想的效果后再利用三角剖分重建叶面三维模型。A large quantity of point cloud data of outdoor trees are obtained by a terrestrial laser scanner （TLS）,with leaf-related point cloud data intercepted from it for blade curve surface fitting and real 3D model construction. The curve surface fitting method for point cloud data scattered in space is mainly studied, and Deluanay triangulation is used to construct a three-dimensional model. Based on the two-dimensional curve surface fitting method of moving least square, in view of the point cloud data scattered in space, a new curve surface fitting method is presented. Through the curve surface fitting of point cloud data based on moving least square, a lead 3D model is reconstructed using Deluanay triangulation after a desired result is obtained.

关 键 词：激光扫描仪 点云数据 曲面拟合 移动最小二乘法 DELAUNAY三角剖分 

分 类 号：TS512[轻工技术与工程—皮革化学与工程]