论文部分内容阅读
针对建(构)筑物点云建模过程中可能碰到的圆柱面拟合问题,本文探讨了一种基于几何特性的圆柱面拟合方法。该方法利用主成分分析法和线性最小二乘法确定拟合初值,再利用非线性最小二乘法循环迭代求解圆柱模型参数,实现圆柱面拟合。它的特点在于误差方程式的建立,本文算法在传统算法的基础上对误差方程稍作改进。实验结果表明,相对于传统方法,采用本文算法计算所得结果的残差与标准差有所降低,精度提高。
In order to solve the cylinder surface fitting problem that may be encountered during the modeling of building point cloud, this paper discusses a cylinder surface fitting method based on geometric characteristics. The method uses principal component analysis and linear least square method to determine the initial value of fitting, and then uses the nonlinear least squares method to iteratively iteratively solve the cylindrical model parameters to achieve cylindrical surface fitting. It is characterized by the establishment of error equations, the algorithm in the traditional algorithm based on the error equation slightly improved. The experimental results show that compared with the traditional method, the residuals and standard deviations of the results calculated by our algorithm are reduced and the accuracy is improved.