Automatic Rail Information Extraction Method Based on Point Cloud Data of 3D Mobile Measurement System

Dongfang ZHANG; Liyuan GU; Yuwen JI; Kangliang WU; Jianfeng LI

doi:10.13890/j.issn.1000-128x.2021.02.018

您当前的位置：

首页 >

文章列表页 >

Automatic Rail Information Extraction Method Based on Point Cloud Data of 3D Mobile Measurement System

Intelligent Technology | 更新时间：2021-12-09

- Automatic Rail Information Extraction Method Based on Point Cloud Data of 3D Mobile Measurement System
- Electric Drive for Locomotives Issue 2, Pages: 114-119(2021)
- 作者机构：
  
  1.株洲时代电子技术有限公司，湖南　株洲　412007
- 作者简介：
- 基金信息：
- DOI：10.13890/j.issn.1000-128x.2021.02.018
  CLC：
扫描看全文
Dongfang ZHANG, Liyuan GU, Yuwen JI, et al. Automatic Rail Information Extraction Method Based on Point Cloud Data of 3D Mobile Measurement System. [J]. Electric Drive for Locomotives (2):114-119(2021)
DOI：

Dongfang ZHANG, Liyuan GU, Yuwen JI, et al. Automatic Rail Information Extraction Method Based on Point Cloud Data of 3D Mobile Measurement System. [J]. Electric Drive for Locomotives (2):114-119(2021) DOI： 10.13890/j.issn.1000-128x.2021.02.018.

摘要

为实现铁路基础设施智能化管理运维，利用3D移动测量系统高效获取高精度点云数据，并对点云数据进行智能化处理。钢轨作为铁路基础设施中最基本的单元，是几何参数计算的基础，对其信息进行自动提取具有重要意义。因此，提出了一种基于3D移动测量系统点云数据的钢轨信息自动提取方法。首先利用点云数据中的角度信息快速实现道床区域的分割，有效减小计算量；然后利用精细栅格划分和动态阈值实现地面点与非地面点的分离；最后利用DBSCAN聚类算法与RANSAC算法完成钢轨点云数据的最终提取。为验证该算法的有效性，以国铁场景钢轨点云数据和隧道场景钢轨点云数据为试验对象，验证结果显示国铁场景和隧道场景的钢轨点云提取准确度分别为96.32%和97.54%，完整度分别为92.14%和94.87%，准确度和完整度均高于90%。试验结果表明：该方法具有操作简单，提取结果准确的优点。

Abstract

In order to realize the intelligent management and maintenance of railway infrastructure, it's necessary to use 3D mobile measurement system to ef fi ciently obtain high-precision point cloud data and intelligently process the point cloud data. As the basic unit of railway infrastructure, rail track is the basis of geometric parameters calculation, so it is of great signi fi cance to automatically segment the rail track. Based on this, an automatic rail information extraction method was proposed based on 3D mobile measurement system point cloud data in this paper. Firstly, the angle information in the point cloud data was used to segment the track bed area, which effectively reduces the amount of calculation. Then, the ground points and non-ground points were separated by using fi ne grid division and dynamic threshold. Finally, DBSCAN clustering algorithm and RANSAC algorithm were used to complete the fi nal extraction of rail point cloud data. In order to verify the effectiveness of the proposed algorithm, rail point cloud data of national railway scene and rail point cloud data of tunnel scene were selected for processing. The rail point cloud extraction accuracy of the railway scene and the tunnel scene were 96.32% and 97.54%, respectively, and the completeness were 92.14% and 94.87%respectively. The accuracy and completeness were both higher than 90%. The experimental results showed that the algorithm has the advantages of simple operation and accurate extraction results.

关键词

移动测量系统点云数据钢轨栅格划分动态阈值轨道交通

Keywords

mobile measurement systempoint cloud datarail trackgrid divisiondynamic thresholdrail transit

references

魁发堂, 石少华, 王龙. 铁路三维地理信息系统构建技术研究[J]. 测绘与空间地理信息, 2016, 39(8): 143-145.

KUI Fatang, SHI Shaohua, WANG Long. Research on 3D modeling and visualization technology of railway lines[J]. Geomatics & Spatial Information Technology, 2016, 39(8): 143-145.

陈隆. 铁路地理数据采集与管理系统建设应用探讨[J]. 勘察科学技术, 2020(4): 45-49.

CHEN Long. Application discussion on railway geographic data acquisition and management system construction[J]. Site Investigation Science and Technology, 2020(4): 45-49.

谭贲. 基于车载激光扫描数据的城市典型地物分类方法研究[D]. 北京: 首都师范大学, 2011.

TAN Ben. Research on city's typical terrain classification based on vehicle-mounted scanning technology[D]. Beijing: Capital Normal University, 2011.

李永强, 盛业华, 刘会云, 等. 基于车载激光扫描的公路三维信息提取[J]. 测绘科学, 2008, 33(4): 23-25.

LI Yongqiang, SHENG Yehua, LIU Huiyun, et al. 3D road information extraction based on vehicle-borne laser scanning[J]. Science of Surveying and Mapping, 2008, 33(4): 23-25.

ELBERINK S O, KHOSHELHAM K, ARASTOUNIA M, et al. Rail track detection and modelling in mobile laser scanner data[C]//ISPRS. ISPRS Conference Laser Scanning 2013. Antalya: ISPRS, 2013(II-5/W2): 223-228. DOI: 10.5194/isprsannals-II-5-W2-223-2013http://doi.org/10.5194/isprsannals-II-5-W2-223-2013.

YANG Bisheng, FANG Lina. Automated extraction of 3-d railway tracks from mobile laser scanning point clouds[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(12), 4750-4761. DOI: 10.1109/JSTARS.2014.2312378http://doi.org/10.1109/JSTARS.2014.2312378.

ARASTOUNIA M, ELBERINK S O. Application of template matching for improving classification of urban railroad point clouds[J]. Sensors, 2016, 16(12): 2112. DOI: 10.3390/s16122112http://doi.org/10.3390/s16122112.

许磊, 李亚辉, 潘成瑜, 等. 基于移动激光点云的轨道中线自动检测[J]. 铁道勘察, 2020, 46(6): 1-6.

XU Lei, LI Yahui, PAN Chengyu, et al. Automatic detection of orbit midline based on moving laser point cloud[J]. Railway Investigation and Surveying, 2020, 46(6): 1-6.

麻卫峰, 王金亮, 王成, 等. 一种基于机载LiDAR点云电力线自动提取方法[J]. 测绘科技与技术学报, 2019, 36(1): 39-44.

MA Weifeng, WANG Jinliang, WANG Cheng, et al. An extraction algorithm of power lines from airborne LiDAR point clouds[J]. Journal of Geomatics Science and Technology, 2019, 36(1): 39-44.

高阳, 王留召, 李明. 车载激光彩色点云的道路标志线提取方法[J]. 测绘通报, 2016(4): 28-32.

GAO Yang, WANG Liuzhao, LI Ming. Study on the extraction methods of road signs based on vehicular color laser point cloud[J]. Bulletin of Surveying and Mapping, 2016(4): 28-32.

Views

下载量

CSCD

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

Development status and suggestions of hydrail technology

Overview on current research status and development trends of hydrogen-powered rail transit

Study on architecture and control technology of multi-stack fuel cell system for rail transit

Research on special gauge of mountain rack railway

Nonlinear compensation-based self-positioning method for rail transit train

Related Author

No data

Related Institution

School of Traffic and Transportation, Beijing Jiaotong University

State Key Lab of Advanced Rail Autonomous Operation, Beijing Jiaotong University

Locomotive & Car Research Institute, China Academy of Railway Sciences Corporation Limited

School of Electrical Engineering, Southwest Jiaotong University

College of Automotive Studies, Tongji University

⁰