基于改进区域生长法的洞室支护锚杆点云提取方法研究

Research on Point Cloud Extraction Method of Cavern Supporting Anchor Based on Improved Region Growing Method

针对施工期抽水蓄能电站地下厂房支护锚杆点云影响数据处理精度的问题,提出了一种基于反射强度改进区域生长算法提取锚杆点云的方法。通过分析锚杆表面曲率及反射强度特征,推导平滑及曲率阈值对提取结果的影响关系,确定基于反射强度的自适应阈值区域生长的改进方法,并通过采集洞室点云及提取支护锚杆点云分析该算法的提取结果。试验表明,该算法效率高,稳定性好,其提取点云簇准确率达98.09%,各簇锚杆的平均点云误提取率仅为2.61%。

Aiming at the problem of the impact of anchor point cloud on the processing accuracy of support anchor cloud data of the underground powerhouse of pumped-storage power station during construction period,a method of extracting the anchor point cloud based on the improved region growing algorithm of reflection intensity is proposed.By analyzing the curvature and reflection intensity characteristics of anchor surface,the influences of the smoothness and curvature threshold on the extraction results are deduced.The improvement on adaptive threshold region growing method based on the reflection intensity is determined,and the extraction results of the algorithm are analyzed by collecting the point cloud of cavern and extracting the point cloud of supporting anchor.The experimental results show that the algorithm has high efficiency and good stability.The accuracy of extracting point cloud clusters is 98.09%,and the average point cloud error extraction rate of each cluster anchor is only 2.61%.