基于改进DBSCAN算法的岩体结构面智能识别方法  

Intelligent identification of rock discontinuities based on an improved DBSCAN algorithm

作　　者：李胜 熊自明 刘一鸣 李志浩 LI Sheng;XIONG Ziming;LIU Yiming;LI Zhihao(State Key Laboratory of Disaster Prevention and Mitigation of Explosion and Impact,Army Engineering University of PLA,Nanjing 210007,Jiangsu,China)

机构地区：[1]陆军工程大学爆炸冲击防灾减灾国家重点实验室,江苏南京210007

出　　处：《隧道与地下工程灾害防治》2022年第2期49-58,共10页Hazard Control in Tunnelling and Underground Engineering

基　　金：国家自然科学基金资助项目(51527810)。

摘　　要：为提高岩体结构面信息获取精度,基于结构面三维点云数据,改进DBSCAN(density-based spatial clustering of applications with noise)密度聚类算法。利用k近邻算法与基于密度比S的评价准则划分出不同密度点云区域以自适应设置参数ε并分析点云法向量,引入法向量夹角阈值T判断属于同组结构面的点集并用相同颜色显示。本研究讨论了参数最优化组合对识别结果的影响,实现优势结构面与产状赤平投影的快速分析。研究结果能够为结构面信息的智能化高效测量提供一种可靠的应用方法。To improve the accuracy of information acquisition,the density-based spatial clustering application with noise(DBSCAN)algorithm was improved based on 3 D point clouds.The k-nearest neighbor algorithm and the evaluation criterion based on density ratio S were used to divide the point cloud regions with different densities in order to set parameterεand analyze the point cloud normal vector adaptive.The angle threshold T of normal vectors was introduced to determine the points belonging to the same plane and the points belonging to the same plane were displayed with the same colour.This paper discussed the influence of different parameter combinations on the identification results and enabled a fast analysis of rock joints.The research results could provide a reliable application method for efficient measurement of discontinuity information.

关 键 词：DBSCAN算法 密度聚类算法 参数最优化 智能识别 岩体结构面 赤平投影 三维点云数据 法向量 

分 类 号：U213.13[交通运输工程—道路与铁道工程]