三维激光雷达点云空间多特征分割

doi:10.11896/jsjkx.210300275

计算机科学 ›› 2022, Vol. 49 ›› Issue (8): 143-149.doi: 10.11896/jsjkx.210300275

• 计算机图形学& 多媒体 • 上一篇下一篇

三维激光雷达点云空间多特征分割

杨文坤, 原晓佩, 陈小锋, 郭睿

西北工业大学自动化学院西安 710129

收稿日期:2021-03-29 修回日期:2021-08-04 发布日期:2022-08-02
通讯作者: 陈小锋(chenxf@nwpu.edu.cn)
作者简介:(ywknpu@163.com)
基金资助:
装备预先研究领域基金(61404130125,61404130118);陕西省自然科学基金(2019JQ-418)

Spatial Multi-feature Segmentation of 3D Lidar Point Cloud

YANG Wen-kun, YUAN Xiao-pei, CHEN Xiao-feng, GUO Rui

School of Automation,Northwestern Polytechnical University,Xi’an 710129,China

Received:2021-03-29 Revised:2021-08-04 Published:2022-08-02
About author:YANG Wen-kun,born in 1996,postgraduate.His main research interests include image processing and lidar remote sensing.
CHEN Xiao-feng,born in 1974,Ph.D,associate professor.His main research interests include traffic information engineering and control,machine vision and embedded system application.
Supported by:
Equipment Pre-research Field Fund (61404130125,61404130118) and Natural Science Foundation of Shaanxi Province,China(2019JQ-418).

摘要/Abstract

摘要： 多线固态激光雷达已成为无人平台环境感知的重要工具,在车载环境建模中得到大量应用。但由于激光雷达分辨率较低、环境噪声敏感、场景复杂等问题,造成场景分割困难。针对实测车载多线激光雷达数据中建筑体点云及植被点云的曲率差异,提出了一种改进的多线激光雷达三维点云的快速分割方法。在曲率分割的基础上,采用加权欧聚类进行二次迭代分割,减少迭代的同时避免陷入局部最优。通过无人平台实测数据采集处理实验和公共数据实验,验证了空间多特征分割方法在激光雷达点云分割方面的有效性;针对场景最终的分割结果,在场景过分割率、欠分割率及正确分割率方面进行了统计,并与传统区域生长分割算法进行了对比分析,结果表明空间多特征分割算法在不同场景的分割中具有较强的适用性及分割准确度。

关键词: 激光雷达, 加权欧氏距离, 建筑体, 曲率分割, 三维点云

Abstract: Multi-layer solid-state lidar has become an important tool for environment perception of unmanned platform,and has been widely used in vehicle-mounted environment modeling.Due to the low resolution of lidar,the sensitivity of environmental noise,and the complexity of the scene,the fast and effective segmentation of the scene becomes a key problem in the real-time environment modeling.In view of the obvious curvature difference between buildings and vegetation in the actual collected point cloud data,this paper proposes an improved fast segmentation method of 3D point cloud based on multi-layer lidar.After the initial segmentation of building facade is realized based on curvature segmentation,the weighted Euclidean clustering is used for the second iterative segmentation of the initial segmented point cloud,which can reduce the iterative process and avoid falling into local optimum.Through the unmanned platform data acquisition and processing experiments and public data experiments,the effectiveness of this method in the segmentation of building and vegetation is verified.According to the final segmentation results of the scene,the over segmentation rate,under segmentation rate and correct segmentation rate of the scene are counted,and compared with the traditional region growing segmentation algorithm.The results show that the algorithm has strong applicability and segmentation accuracy in different scenes.

Key words: 3D point cloud, Building, Curvature segmentation, Lidar, Weighted Euclidean distance

中图分类号:

TP79

杨文坤, 原晓佩, 陈小锋, 郭睿. 三维激光雷达点云空间多特征分割[J]. 计算机科学, 2022, 49(8): 143-149. https://doi.org/10.11896/jsjkx.210300275

YANG Wen-kun, YUAN Xiao-pei, CHEN Xiao-feng, GUO Rui. Spatial Multi-feature Segmentation of 3D Lidar Point Cloud[J]. Computer Science, 2022, 49(8): 143-149. https://doi.org/10.11896/jsjkx.210300275

参考文献

[1]MUSIALSKI P,WONKA P,ALIAGA D G,et al.A Survey ofUrban Reconstruction [J].Computer Graphics Forum,2013,32(6):146-177.
[2]HE Y R,ZHENG Y M,PAN H P,et al.Real Three-dimensional Modeling and Application of Complex Construction based on the Point Cloud Data [J].Remote Sensing Technology and Application,2016,31(6):1091-1099.
[3]ZHANG P Z,ZHANG H Y.A Review of Features and Labels Dimensionality Reduction Methods of Multi Label Data[J].Journal of Chongqing Technology and Business University(Na-tural Science Edition),2020,37(5):23-29.
[4]HONG S X,WANG J X.Geomatics S O.Aerial LiDAR Building Point-cloud Extraction Algorithm Combining OTSU and Iterative TIN[J].Remote Sensing Information,2018,33(6):79-85.
[5]YANG Y,YANG G,ZHENG T,et al.Feature extraction me-thod based on 2.5-dimensions lidar platformfor indoor mobile robots localization[C]//2017 IEEE International Conference on Cybernetics and Intelligent Systems(CIS) and Automation and Mechatronics(RAM).2017:28-34.
[6]WEI Z,YANG B Q,LI Q Q.Automated extraction of building footprints from mobile LIDAR point clouds[J].Journal of Remote Sensing,2012,16(2):286-296.
[7]ZHU J T,WANG L,ZHAO Z.Point cloud segmentation ofcomplex building roof based on region growing algorithm[J].Remote Sensing of Land and Resources,2019,31(4):20-25.
[8]HU H L,LI Z,JIN X G,et al.Curve Skeleton Extraction From 3D Point Clouds Through Hybrid Feature Point Shifting and Clustering[J].Computer Graphics Forum,2020,39(6):111-132.
[9]AHMED M,RAIHAN SERAJ R,ISLAM S M S.The k-means Algorithm:A Comprehensive Survey and Performance Evaluation [J].Electronics,2020,9(8):1295.
[10]SUN S L,YANG Q,ZHANG Y M.A Parameter Simplification Convolutional Neural Network Model for Image Segmentation[J].Journal of Chongqing University of Technology(Natural Science),2021,35(3)145-151.
[11]WU Y L,LI Y Q.Convolutional Network Based Pathological Nucleus Segmentation[J].Journal of Chongqing Technology and Business University(Natural Science Edition),2019,36(3):67-71.
[12]QIU Z,ZHUANG Y,YAN F,et al.RGB-DI images and fullconvolution neural network-based outdoor scene understanding for mobile robots[J].IEEE Transactions on Instrumentation and Measurement,2018,68(1):27-37.
[13]KHAN I,LUO Z,HUANG J Z,et al.Variable Weighting inFuzzy K-means Clustering To Determine the Number of Clusters[J].IEEE Transactions on Knowledge and Data Enginee-ring,2020,32(9):1938-1853.
[14]HE Y B,CHEN R L,WU K,et al.Point cloud simplificationmethod based on K-means clustering[J].Progress in Laser and Optoelectronics,2019,56(9):96-99.
[15]DENG D S.Application of DBSCAN Algorithm in Data Sam-pling[J].Journal of Physics:Conference Series,2020,1617(1):012088.
[16]CUI Z.Extracting facade information of the side hall of the Forbidden City by European clustering [J].Science and Technology Innovation,2020(18):75-78.
[17]BORCS A,NAGY B,BENEDEK C.Instant Object Detection in Lidar Point Clouds[J].IEEE Geoscience & Remote Sensing Letters,2017,14(7):992-996.
[18]SHI H E,SUN X Y,HUANG J H.Dehazing Algorithm for Remote Sensing Image Optimization Based on Curvature Filtering[J].Acta Photonica Sinica,2021,50(2):0210003.
[19]LI S L.Research of Pulsed Laser Ranging system[D].Guilin:Guilin Universoty of Electronic Technology,2015.
[20]ZHANG X S,HUA S G.Surface Reconstruction from Point Cloud Combining Plane Projection and Region-growing[J].Mechanical&Electrical Englneering Technology,2020,49(6):23-24,153.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

三维激光雷达点云空间多特征分割

Spatial Multi-feature Segmentation of 3D Lidar Point Cloud

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

Metrics

本文评价

推荐阅读 0

[1]	田野, 陈宏巍, 王法胜, 陈兴文. 室内移动机器人的SLAM算法综述 Overview of SLAM Algorithms for Mobile Robots 计算机科学, 2021, 48(9): 223-234. https://doi.org/10.11896/jsjkx.200700152
[2]	李广敬, 鲍泓, 徐成. 一种基于3D激光雷达的实时道路边缘提取算法 Real-time Road Edge Extraction Algorithm Based on 3D-Lidar 计算机科学, 2018, 45(9): 294-298. https://doi.org/10.11896／j.issn.1002-137X.2018.09.049
[3]	秦绪佳,陈楼衡,谭小俊,郑红波,张美玉. 结构光视觉三维点云逐层三角网格化算法 Layer by Layer Triangulation Algorithm for 3D Point Clouds from Structured Light Vision 计算机科学, 2016, 43(Z11): 383-387. https://doi.org/10.11896/j.issn.1002-137X.2016.11A.088
[4]	张志刚,孙立才,汪沛. 基于激光扫描技术的行人检测方法研究 Research on Pedestrian Detection Method Based on Laser Scanning 计算机科学, 2016, 43(7): 328-330. https://doi.org/10.11896/j.issn.1002-137X.2016.07.061
[5]	周剑云,王丽珍,杨增芳. 基于加权欧氏距离的空间Co-location模式挖掘算法研究 Algorithm of Mining Spatial Co-location Patterns Based on Weighted Euclidean Distance 计算机科学, 2014, 41(Z6): 425-428.
[6]	王钤,郭丽,石航飞. 基于激光雷达距离图像的道路边缘检测与跟踪 Detecting and Tracking Edges of the Lane Based on Range Images 计算机科学, 2013, 40(Z11): 405-408.
[7]	游安清,韩晓言,李世平,闫召进. 激光点云中输电线拟合与杆塔定位方法研究 Transmission Lines Fitting and Towers Positioning in LiDAR Point Cloud 计算机科学, 2013, 40(4): 298-300.