吉林大学学报(工学版) ›› 2022, Vol. 52 ›› Issue (8): 1918-1925.doi: 10.13229/j.cnki.jdxbgxb20220263

• 通信与控制工程 • 上一篇    

基于超体素双向最近邻距离比的点云配准方法

李雪梅1,2(),王春阳1,3(),刘雪莲3,施春浩1,李国瑞1   

  1. 1.长春理工大学 电子信息工程学院，长春 130022
    2.白城师范学院 机械与控制工程学院，吉林 白城 137000
    3.西安工业大学 西安市主动光电成像探测技术重点实验室，西安 710021
  • 收稿日期:2022-03-19 出版日期:2022-08-01 发布日期:2022-08-12
  • 通讯作者: 王春阳 E-mail:lixuemei556677@163.com;wangchunyang19@163.com
  • 作者简介:李雪梅（1973-），女，副教授，博士. 研究方向：光电信息处理. E-mail：lixuemei556677@163.com
  • 基金资助:
    吉林省教育厅科学技术研究重点规划项目(JJKH20220014KJ)

Point cloud registration method based on supervoxel bidirectional nearest neighbor distance ratio

Xue-mei LI1,2(),Chun-yang WANG1,3(),Xue-lian LIU3,Chun-hao SHI1,Guo-rui LI1   

  1. 1.School of Electronic and Information Engineering，Changchun University of Science and Technology，Changchun 130022，China
    2.School of Mechanical and Control Engineering，Baicheng Normal University，Baicheng 137000，China
    3.Xi'an Key Laboratory of Active Photoelectric Imaging Detection Technology，Xi'an Technological University，Xi'an 710021，China
  • Received:2022-03-19 Online:2022-08-01 Published:2022-08-12
  • Contact: Chun-yang WANG E-mail:lixuemei556677@163.com;wangchunyang19@163.com

RICH HTML

 1   

PDF (PC)

265

摘要：

针对点云数据冗余和配准精度低的问题，提出了一种基于超体素的双向最近邻距离比匹配的配准方法。首先，利用超体素提取了具有稳定结构的目标特征点，同时提出了利用点云厚度分层进行非迭代的阈值去噪方法；然后，利用FPFH进行特征描述，提出了双向最近邻距离比方法对点云进行了初始配准；最后，提出了基于双级阈值的点云精确配准方法。采用标准数据库模型进行仿真分析，验证了算法的有效性。结果表明：本方法能有效剔除漂移噪声体素，配准精度高，鲁棒性强。与其他方法对比，在配准时间相近时，本文算法的配准精度提高74.2%；在噪声占比为6%和10%时，配准精度均提高67%以上。

关键词: 信号与信息处理, 点云配准, 超体素, 点云分层去噪, 双向最近邻距离比, 双级阈值配准

Abstract:

Aiming at the problems of redundancy and low registration accuracy of the point cloud data， a bidirectional nearest neighbor distance ratio registration method based on supervoxel is proposed in this paper. Firstly， the target feature points with stable structure are extracted by the supervoxel， and a non-iterative threshold denoising method based on point cloud thickness stratification is proposed； then， using FPFH for feature description， and a bidirectional nearest neighbor distance ratio method is proposed to register the point cloud； finally， an accurate point cloud registration method based on two-level threshold is proposed. The standard database model is used for simulation analysis to verify the effectiveness of the algorithm. The results show that the proposed method in this paper can effectively eliminate the drift noise voxels， and the algorithm has high registration accuracy and strong robustness. Compared with other methods， when the registration time is close， the registration accuracy of this algorithm is improved by 74.2%； When the noise ratio is 6% and 10%， the registration accuracy is improved by more than 67%.

Key words: signal and information processing, point cloud registration, supervoxel, point cloud layered denoising, bidirectional nearest neighbor distance ratio, two-level threshold registration

中图分类号: 

  • G202

图1

超体素生成过程"

图2

体素云分层示意图"

图3

FPFH直方图"

图4

初始配准"

图5

双级阈值配准"

表1

配准参数"

参数名称参数值参数名称参数值
体素分辨率0.001距离比阈值0.985
网格分辨率0.008初配迭代30
法向量阈值0.8初配搜索半径0.05
邻域值10网格化去噪搜索半径0.0015
阈值噪声系数0.9成像横向分辨率0.1
测距精度0.1

表2

配准时间对比"

分组算法总时间/s误差
1FPFH+双向 最近邻距离比 初配+ICP77.69071.4281×10-33
FPFH+双向 最近邻距离比 初配+双级 阈值精配76.89233.6786×10-34
2FPFH+单向 最近邻距离初配19.94594.2109×10-2
FPFH+双向最 近邻距离比初配72.66512.4472×10-4

图6

无噪声下配准结果对比"

图7

噪声下配准结果对比"

表3

噪声下配准精度对比"

算 法噪声占比/%误 差
ICP06.3903×10-4
61.2049×10-4
101.1139×10-4
FPFH+ICP01.7161×10-33
61.5233×10-3
102.3226×10-3
FPFH+双级阈值精配03.7351×10-34
65.2333×10-4
105.6391×10-4
本文算法06.5561×10-5
63.8025×10-5
103.5942×10-5
1 党相卫, 秦斐, 卜祥玺, 等. 一种面向智能驾驶的毫米波雷达与激光雷达融合的鲁棒感知算法[J]. 雷达学报, 2021, 10(4): 622-631.
Dang Xiang-wei, Qin Fei, Bu Xiang-xi,et al. A robust perc-eption algorithm based on a radar and lidar for intelligent driving[J]. Journal of Radars, 2021, 10(4): 622-631.
2 王岩, 王飞, 王挺峰, 等. 基于自适应阈值的阵列激光三维点云配准[J]. 物理学报, 2016, 65(24): 271-281.
Wang Yan, Wang Fei, Wang Ting-feng, et al. Laser array imaging point cloud registration based on adaptive threshold[J].Acta Physica Sinica, 2016, 65(24): 271-281.
3 王庆闪, 张军, 刘元盛, 等. 基于NDT与ICP结合的点云配准算法[J]. 计算机工程与应用, 2021, 56(7): 88-95.
Wang Qing-shan, Zhang Jun, Liu Yuan-sheng, et al. Point cloud registration algorithm based on combination of NDT and ICP[J]. Computer Engineering and Applications, 2021, 56(7): 88-95.
4 Chen C, Li Y, Wang W. A 3D sequential LiDAR data registration method for unmanned ground vehicle[J]. Applied Mechanics and Materials, 2014, 664: 365-370.
5 闫利, 索一凡, 曹亮. 车载激光雷达点云与全景图像的配准[J]. 测绘科学, 2016, 41(4): 113-123.
Yan Li, Suo Yi-fan, Cao Liang. Registration of vehicle-bo-rne LiDAR point cloud and panoramic image[J]. Science of Surveying and Mapping, 2016, 41(4): 113-123.
6 He Y Q, Mei Y G.An efficient registration algo-rithm based on spin image for LiDAR 3D point cloud models[J]. Neurocomputing, 2015, 151: 354-363.
7 Zong W P, Li M L, Zhou Y L, et al. A fast and accurate planar-feature-based global scan registration method[J].IEEE Sensors Journal, 2019, 19(24): 12333-12345.
8 陈学伟,朱耀麟,武桐, 等. 基于SAC-IA和改进ICP算法的点云配准技术[J]. 西安工程大学学报, 2017, 31(3): 395-401.
Chen Xue-wei, Zhu Yao-lin, Wu Tong, et al. The point cloud registration technology based on SAC-IA and improved ICP[J]. Journal of Xi'an Polytechnic University, 2017, 31(3): 395-401.
9 赵海鹏,习晓环,王成, 等.基于车载激光扫描数据的城区道路自动提取[J]. 中国科学院大学学报, 2018, 35(6): 782-787.
Zhao Hai-peng, Xi Xiao-huan, Wang Cheng, et al. Automatic extraction of urban road information based on mobile laser scanning data[J]. Journal of University of Chinese Academy of Sciences, 2018, 35(6): 782-787.
10 王新竹,李骏,李红建, 等.基于三维激光雷达和深度图像的自动驾驶汽车障碍物检测方法[J]. 吉林大学学报: 工学版, 2016, 46(2): 360-365.
Wang Xin-zhu, Li Jun, Li Hong-jian, et al. Obstacle detection based on 3D laser scanner and range image for intelligent vechicle[J]. Journal of Jilin University(Engineering and Technology Edition), 2016, 46(2): 360-365.
11 宗长富, 文龙, 何磊. 基于欧几里得聚类算法的三维激光雷达障碍物检测技术[J]. 吉林大学学报: 工学版, 2020, 50(1): 107-113.
Zong Chang-fu, Wen Long, He Lei. Object detection based on Euclidean clustering algorithm with 3D laser scanner[J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(1): 107-113.
12 吴永兴. 利用超体素的车载激光点云杆状目标的提取[J]. 测绘地理信息, 2021, 46(4): 77-81.
Wu Yong-xing. Extracting pole-like objects in mobile laser scanning data with supervoxels[J]. Journal of Geomatics, 2021, 46(4): 77-81.
13 Yu Y T, Li J, Guan H Y, et al. Automated extraction of urban road facilities using mobile laser scanning data[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(4): 2167-2181.
14 赵成伟, 吴云东, 蔡国榕, 等. 基于超体素的LiDAR点云粘连目标分割算法[J]. 集美大学学报: 自然科学版, 2017, 22(1): 73-80.
Zhao Cheng-wei, Wu Yun-dong, Cai Guo-rong, et al. Research of segmentation algorithm for LiDAR point cloud based on supervoxel[J]. Journal of Jimei University(Natural Science), 2017, 22(1): 73-80.
15 Li Y C, Liu Y, Sun R, et al. Multi-view point cloud registration with adaptive convergence threshold and its application on 3D model retrieval[J]. Multimedia Tools and Applications, 2020, 79: 14793-14810.
16 Zhou S T, Liu X L, Wang C Y, et al. Non-iterative denoising algorithm based on a dual threshold for a 3D point cloud[J]. Optics and Lasers in Engineering, 2020, 126 :105921.
17 陈春旭,漆钰晖,朱一帆, 等. ICP配准算法的影响因素及评价指标分析[J]. 导航定位与授时, 2018, 5(5): 67-72.
Chen Chun-xu, Qi Yu-hui, Zhu Yi-fan, et al. The analysis of influence factors and evaluation indexes on ICP algorithm[J]. Navigation Positioning Timing,2018, 5(5): 67-72.
18 Meng Y, Zhang H. Registration of point clouds using sample-sphere and adaptive distance restriction[J]. Visual Computer, 2011, 27(6-8): 543-553.
[1] 李雪梅,王春阳,刘雪莲,谢达. 基于SESTH的线性调频连续波激光雷达信号时延估计[J]. 吉林大学学报(工学版), 2022, 52(4): 950-958.
[2] 窦慧晶,丁钢,高佳,梁霄. 基于压缩感知理论的宽带信号波达方向估计[J]. 吉林大学学报(工学版), 2021, 51(6): 2237-2245.
[3] 金心宇,谢慕寒,孙斌. 基于半张量积压缩感知的粮情信息采集[J]. 吉林大学学报(工学版), 2021, 51(1): 379-385.
[4] 郭立民,陈鑫,陈涛. 基于AlexNet模型的雷达信号调制类型识别[J]. 吉林大学学报(工学版), 2019, 49(3): 1000-1008.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 《吉林大学学报(工学版)》编辑部
地址：长春市人民大街5988号　电话：+86-431-85095297　传真：+86-431-85094074　E-mail: xbgxb@jlu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发