论文部分内容阅读
为了提高变形监测中地铁隧道断面点截取的效率,文章提出了基于kd-tree和法向量估计的局部点云简化方法,对BaySAC算法的三维激光点云二次参数曲面拟合方法进行改进:利用kd-tree建立点云数据的空间拓扑关系,计算出每个数据点的k邻域;然后使用平面拟合方法获取法矢量;最后根据点云数据法矢量变化程度,采用法矢量自适应得到压缩后的点云数据。实验证明该方法既能较大程度地简化点云,简化结果比较均匀,又具有不破坏细小特征的特点,进一步改进了BaySAC算法的二次参数曲面拟合方法。
In order to improve the efficiency of interception of cross-section points of metro tunnel in deformation monitoring, a simplified method of local point cloud based on kd-tree and normal vector estimation is proposed to improve Bayesian algorithm for 3D parametric surface fitting of laser point cloud. kd-tree to establish the spatial topology of point cloud data to calculate the k-neighborhood of each data point; then use the plane fitting method to obtain the normal vector; Finally, according to the degree of vector change of point cloud data method, After the point cloud data. Experiments show that this method not only simplifies the point cloud to a great extent, but also simplifies the result more evenly and has the characteristics of not destroying the fine features. The quadratic parametric surface fitting method of BaySAC algorithm is further improved.