摘要
现实中的三维激光点云数据中存在大量圆柱几何体,在实际操作中,由于测量环境的干扰性和实验仪器的局限性,使圆柱体测量数据的拟合精度受到影响。本文在随机抽样一致性算法的基础上研究一种基于多元回归随机抽样一致性算法的圆柱拟合方法,利用MATLAB编程实现对圆柱的轴线方向向量和半径拟合求解,通过设置仿真数据和实测数据实验与最小二乘及整体最小二乘拟合结果进行比较,结果表明本文算法对于圆柱拟合过程中的参数解算具有较好的鲁棒性。
There are a large number of cylindrical geometry in the reality of 3 D laser point cloud data,and in practice,due to the interference of the measurement environment and the limitation of experimental instruments,the fitting accuracy of measurement data is affected.Based on the random sample consensus algorithm,this paper studies a cylindrical fitting method by using the multi-regression,and uses MATLAB programming to solve the parameters of axis direction vector and radius of the fitting cylinder.The comparison results between the least squares and total least squares on simulation and real examples show that this method is robust in cylindrical fitting.
作者
顾冲
曹爽
吉宸佳
张杰
郭奕
Gu Chong;Cao Shuang;Ji Chenjia;Zhang Jie;Guo Yi(School of Remote Sensing&Geomatics Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China)
出处
《工程勘察》
2022年第5期47-52,72,共7页
Geotechnical Investigation & Surveying
基金
国家级大学生创新训练项目(201910300036Z)。
