[1]马明刚,郑德华,潘月梁,等.基于曲率频数统计的支护锚杆点云自适应提取方法研究[J].南京师大学报(自然科学版),2022,45(03):27-34.[doi:10.3969/j.issn.1001-4616.2022.03.005]
　Ma Minggang,Zheng Dehua,Pan Yueliang,et al.Study on Adaptive Extraction Method of Rod-like Point Cloud Based on Curvature Frequency Statistics[J].Journal of Nanjing Normal University(Natural Science Edition),2022,45(03):27-34.[doi:10.3969/j.issn.1001-4616.2022.03.005]
点击复制

基于曲率频数统计的支护锚杆点云自适应提取方法研究()
分享到：

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
第45卷
期数:
2022年03期
页码:
27-34
栏目:
物理学
出版日期:
2022-09-15

文章信息/Info

Title:
Study on Adaptive Extraction Method of Rod-like Point Cloud Based on Curvature Frequency Statistics
文章编号:
1001-4616(2022)03-0027-08
作者:
马明刚1; 郑德华2; 潘月梁1; 李思远2; 张 兵1; 胡 创2
(1.浙江宁海抽水蓄能有限公司,浙江 宁海 315600)(2.河海大学地球科学与工程学院,江苏 南京 210098)
Author(s):
Ma Minggang1; Zheng Dehua2; Pan Yueliang1; Li Siyuan2; Zhang Bing1; Hu Chuang2
(1.Zhejiang Ninghai Pumped Storage Co.,Ltd,Ninghai 315600,China)(2.School of Earth Sciences and Engineering,Hohai University,Nanjing 210098,China)
关键词:
三维激光扫描; 施工洞室; 锚杆点云识别; 曲率阈值法; 曲率频数直方图
Keywords:
3D laser scanning; construction cavern; anchor point cloud identification; curvature threshold method; curvature frequency histogram
分类号:
P258
DOI:
10.3969/j.issn.1001-4616.2022.03.005
文献标志码:
A
摘要:
针对抽水蓄能电站地下厂房施工期拱顶的支护锚杆,提出了一种从激光扫描采集的洞室密集点云中提取支护锚杆点云方法. 设计了锚杆点云初识别和锚杆点云精提取的洞室支护锚杆点云提取过程. 首先对获取的密集点云进行保留支护锚杆轮廓特征的降采样,通过建立局部坐标系的点云坐标分布特征,结合支护锚杆的实际结构参数判定,初识别存在支护锚杆疑似点云的区域; 再采用曲率阈值法精提取锚杆结构点云,分类提取支护锚杆部位点云. 通过实际工程采集的支护锚杆点云提取处理试验,结果表明:设计的支护锚杆点云簇初识别方法准确率为100%; 采用基于曲率频数直方图曲线拟合的曲率阈值确定方法具有较好的适用性,能够将洞室中 4类特征支护锚杆点云从拱顶粗糙表面的密集点云中精确地提取出来,可为施工洞室点云精确配准和变形分析提供优质可靠的数据源.
Abstract:
Aiming at the supporting anchor of top cavern during the construction period of underground plant of the pumped storage power station,a method of extracting supporting anchor point cloud from dense point cloud of the cavern collected by laser scanning was proposed. The extraction process of the cave support anchor point cloud extraction is designed for the initial identification of anchor point clouds and the extraction of anchor point cloud essence. Firstly,the acquired dense point cloud data was down-sampled to retain the contour characteristics of supporting anchor,and the point cloud coordinate distribution characteristics of local coordinate system were established. Combined with the actual structural parameters of supporting anchor,the area where suspected point cloud of supporting anchor was initially identified; then the curvature threshold method was used to accurately identify the point cloud of anchor structure,and the point cloud of supporting anchor was classified and extracted. Through the extraction and processing test of supporting anchor point cloud collected by actual project,the results show that supporting anchor point cloud identification method has an accuracy of 100% and the curvature threshold determination method based on curve frequency histogram curve fit has good applicability,which can accurately extract the supporting anchor point cloud with 4 types of characteristics in cavern from the dense point cloud on the rough surface of top cavern,which can provide high-quality and reliable data sources for the accurate registration and deformation analysis of construction cavern point cloud.

参考文献/References:

[1]李海波,杨兴国,赵伟,等. 基于三维激光扫描的隧洞开挖衬砌质量检测技术及其工程应用[J]. 岩石力学与工程学报,2017,36(S1):3456-3463.
[2]范泯进,罗浩,朱燕梅,等. 基于三维激光扫描技术的水工隧洞变形量高精度计算方法研究[J]. 水利规划与设计,2019(8):79-81.
[3]刘元元. 基于点云数据的锚杆地物提取方法[J]. 工程勘察,2022,50(4):62-66.
[4]杨必胜,魏征,李清泉,等. 面向车载激光扫描点云快速分类的点云特征图像生成方法[J]. 测绘学报,2010,39(5):540-545.
[5]王鹏,刘如飞,马新江. 车载激光点云中锚杆目标自动提取[J]. 遥感信息,2017,32(5):52-56.
[6]黄礼辉,黄子懿. 一种基于数学形态学的点云地物提取方法[J]. 铁道勘察,2018,44(2):26-29.
[7]刘哲. 基于水下三维点云的管道检测定位算法研究[D]. 哈尔滨:哈尔滨工程大学,2020. DOI:10.27060/d.cnki.ghbcu.2020.001332.
[8]刘哲,张万远,黄晓霜,等. 海底管道检测与三维点云重建算法[J]. 海洋测绘,2019,39(3):36-40.
[9]JIA D F,ZHANG W P,WANG Y H,et al. A new approach for cylindrical steel structure deformation monitoring by dense point clouds[J]. Remote sensing,2021,13(12):2263-2279.
[10]吴永兴. 利用超体素的车载激光点云锚杆目标的提取[J]. 测绘地理信息,2021,46(4):77-81.
[11]李鹏鹏. 基于车载LiDAR数据的道路场景典型锚杆地物分类和动态分析[D]. 焦作:河南理工大学,2020. DOI:10.27116/d.cnki.gjzgc.2020.000161.
[12]SINGH S K,RAVAL S,BANERJEE B. Roof bolt identification in underground coal mines from 3D point cloud data using local point descriptors and artificial neural network[J]. International journal of remote sensing,2020,42(1):367-377.

备注/Memo

备注/Memo:
收稿日期:2022-05-27.
基金项目:国网新源控股(水电)有限公司科技项目(SGXYKJ-2020-079).
通讯作者:郑德华,博士,副教授,研究方向:三维激光扫描数据处理. E-mail:zheng_dehua@163.com
更新日期/Last Update: 2022-09-15