基于三维激光点云的隧道开挖岩体结构面识别与信息提取  

Identification and Information Extraction of Rock Structure Surface of Tunnel Excavation Based on 3D Laser Point Cloud

作　　者：孙文志 杨辉 郭景生 李锴 

机构地区：[1]中铁九局集团第五工程有限公司,四川 成都 [2]西南交通大学土木工程学院,四川 成都

出　　处：《土木工程》2022年第5期690-698,共9页Hans Journal of Civil Engineering

摘　　要：岩体结构面几何参数是评价岩体稳定性与渗流特征的重要研究基础。在隧道施工中,识别开挖面岩体结构并分析为后续施工提供了重要参考价值。采用三维激光扫描技术,获取隧道开挖面岩体点云数据,采用球面投影,对投影数据进行三角剖分,得到隧道施工开挖面三角剖分模型,利用结构面产状特征值和距离特征值,采用动态聚类算法实现了岩体结构面分类与识别,并进行结构面产状和面积信息计算。以本方法对某隧道三维激光扫描点云数据进行测试,结合人工测量数据进行比较,结果表明,采用本方法能够快速对岩体结构面进行分组识别和信息提取。The geometric parameters of the rock structure face are an important research basis for evaluating the stability and seepage characteristics of rock masses. In tunnel construction, the identification and analysis of the rock structure of the excavation face provides an important reference value for subsequent construction. A 3D laser scanning technique is used to obtain the point cloud data of the rock face of the tunnel excavation, a spherical projection is used to triangulate the projection data, a triangulation model of the tunnel excavation is obtained, and the structural face is classified and identified using a dynamic clustering algorithm using the structural face eigenvalues and distance eigenvalues, and the structural face yield and area information are calculated. The method was tested on 3D laser scanned point cloud data of a tunnel and compared with manual measurement data. The results show that the method can be used to quickly identify and extract information from the structural faces of rock masses.

关 键 词：三维激光扫描技术 点云 特征值 动态聚类算法 

分 类 号：U455[建筑科学—桥梁与隧道工程]