一种基于点云场景分割与改进配准算法的物体位姿估计方法

doi:10.3901/JME.2023.22.176

机械工程学报 ›› 2023, Vol. 59 ›› Issue (22): 176-185.doi: 10.3901/JME.2023.22.176

扫码分享

一种基于点云场景分割与改进配准算法的物体位姿估计方法

张瑾贤¹, 吴晓峰², 叶才铭², 杨吉祥¹, 丁汉¹

1. 华中科技大学数字制造装备与技术国家重点实验室武汉 430074;
2. 中国航发南方工业有限公司株洲 412002

收稿日期:2022-11-05 修回日期:2023-04-15 出版日期:2023-11-20 发布日期:2024-02-19
通讯作者: 杨吉祥(通信作者)，男，1987年出生，博士，副教授，博士研究生导师。主要研究方向为机器人和数控加工技术与装备。E-mail：jixiangyang@hust.edu.cn
作者简介:张瑾贤，男，1998年出生。主要研究方向为机器人和三维点云测量。E-mail：jinxian@hust.edu.cn
基金资助:
国家重点研发计划(2020YFB1710400)、国家自然科学基金(52122512，52188102)和湖北省自然科学基金(2021CFA075)资助项目。

Object Pose Estimation Method Based on Point Cloud Scene Segmentation and Improved Registration

ZHANG Jinxian¹, WU Xiaofeng², YE Caiming², YANG Jixiang¹, DING Han¹

1. State Key Laboratory of Digital Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan 430074;
2. Aecc South Industry Company Limited, Zhuzhou 412002

Received:2022-11-05 Revised:2023-04-15 Online:2023-11-20 Published:2024-02-19

摘要/Abstract

摘要： 为了解决低分辨率深度相机获取复杂场景下物体精确位姿信息困难的问题，提出一种基于点云场景分割与改进配准算法的物体位姿估计方法。首先，提出采用结构光三维扫描仪来扫描制作目标物体模板，以消除由理论模型直接生成模板所带来的差异性。而后，提出了一种基于两步法的物体分割方式，能够快速准确地完成场景点云中目标物体的分割。最后，提出一种结合法线夹角约束与邻域数约束的关键点提取算法，能够有效提取模板与场景点云中具有大曲率特征且非噪声的关键点，紧接着在关键点处计算 FPFH 特征描述，并基于随机采样一致性完成物体粗配准与初始位姿估计。为提高位姿估计精度，进一步采用带法线夹角约束的改进 ICP 算法，完成物体初始位姿估计的精确修正。通过试验对所提方法进行了验证，对比现有基于点云配准的位姿估计方法，位姿估计误差明显减小，有效证明了所提方法的可行性。

关键词: 位姿估计, 场景分割, 点云配准, ICP算法, 特征约束

Abstract: To solve the difficulty of obtaining accurate position and pose information of objects in complex scenes with low resolution depth camera, an object pose estimation method based on point cloud scene segmentation and improved registration algorithm is proposed. Firstly, a structured light scanning 3D scanner is proposed to make the template of the object, to eliminate the difference caused by the template directly generated by the theoretical model. Then, an object segmentation method based on two-step method is proposed, which can quickly and accurately complete the segmentation of the object in the scene cloud. Finally, a key point extraction algorithm that combines the normal angle constraint and the neighborhood number constraint is proposed, which can effectively extract the key points with large curvature characteristics and non-noise in the template and scene point clouds, and then calculate FPFH description at the key points, and complete object coarse registration and initial pose estimation based on random sampling consistency. To improve the accuracy of pose estimation, an improved ICP algorithm with normal angle constraint is adopted to complete the accurate correction of the initial pose estimation. The proposed method is verified by experiments. Compared with the existing pose estimation methods based on point cloud registration, the error of pose estimation is significantly reduced, which effectively proves the feasibility of the proposed method.

Key words: pose estimation, scene segmentation, point cloud registration, ICP algorithm, feature constraints

中图分类号:

TP391

张瑾贤, 吴晓峰, 叶才铭, 杨吉祥, 丁汉. 一种基于点云场景分割与改进配准算法的物体位姿估计方法[J]. 机械工程学报, 2023, 59(22): 176-185.

ZHANG Jinxian, WU Xiaofeng, YE Caiming, YANG Jixiang, DING Han. Object Pose Estimation Method Based on Point Cloud Scene Segmentation and Improved Registration[J]. Journal of Mechanical Engineering, 2023, 59(22): 176-185.

参考文献

[1] XU Chi, CHEN Jiale, YAO Mengyang, et al. 6DoF pose estimation of transparent object from a single RGB-D image[J]. Sensors, 2020, 20(23):6790.
[2] LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.
[3] BAY H, TUYTELAARS T, VAN GOOL L. SURF:Speeded up robust features[C]//9th European Conference on Computer Vision, May 07-13, 2006, Graz, Austria. Springer-Verlag, 2006:404-417.
[4] 刘然,范秀敏,尹旭悦,等. AR辅助装配中基体及零件位姿组合估计方法[J]. 机械设计与研究, 2018, 34(6):119-125. LIU Ran, FAN Xiumin, YIN Xuyue, et al. Pose combination estimation method of matrix and parts in AR assisted assembly[J]. Machine Design & Research, 2018, 34(6):119-125.
[5] HINTERSTOISSER S, HOLZER S, CAGNIART C, et al. Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes[C]//2011 International Conference on Computer Vision, November 6-13, 2011, Barcelona, Spain. IEEE, 2011:858-865.
[6] HODAN T, ZABULIS X, LOURAKIS M, et al. Detection and fine 3D pose estimation of texture-less objects in RGB-D images[C]//IEEE/RSJ International Conference on Intelligent Robots & Systems, September 28, 2015, Hamburg, Germany. IEEE, 2015:4421-4428.
[7] 王梓,孙晓亮,李璋,等. 基于Transformer模型的卫星单目位姿估计方法[J]. 航空学报, 2022, 43(5):325298. WANG Zi, SUN Xiaoliang, LI Zhang, et al. Satellite monocular pose estimation method based on transformer model[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(5):325298.
[8] CHAI Ziqi, BI Sheng, GUO Zheng, et al. A new method for fast detection and pose estimation of texture-less industrial parts[C]//25th International Conference on Mechatronics and Machine Vision in Practice (M2VIP), November 20-22, 2018, Stuttgart, Germany. IEEE, 2018:79-84.
[9] DROST B, ULRICH M, NAVAB N, et al. Model globally, match locally:Efficient and robust 3D object recognition[C]//2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 13-18, 2010, San Francisco, CA, USA. IEEE, 2010:998-1005.
[10] LI Deping, WANG Hanyun, LIU Ning, et al. 3D object recognition and pose estimation from point cloud using stably observed point pair feature[J]. IEEE Access, 2020, 8:44335-44345.
[11] 鲁荣荣,朱枫,吴清潇,等. 基于增强型点对特征的三维目标识别方法[J]. 光学学报, 2019, 39(8):245-254. LU Rongrong, ZHU Feng, WU Qingxiao, et al. 3D object recognition based on enhanced point pair feature[J]. Acta Optica Sinica, 2019, 39(8):245-254.
[12] MIAO Yanzi, LIU Yang, MA Hongbin, et al. The pose estimation of mobile robot based on improved point cloud registration[J]. International Journal of Advanced Robotic Systems, 2016, 13(2):52.
[13] HU Haopeng, GU Weikun, YANG Xiansheng, et al. Fast 6D object pose estimation of shell parts for robotic assembly[J]. The International Journal of Advanced Manufacturing Technology, 2021, 118:1383-1396.
[14] SONG Kaitai, WU Chenghei, JIANG Sinyi. CAD-based pose estimation design for random bin picking using a RGB-D camera[J]. Journal of Intelligent & Robotic Systems, 2017, 87(3-4):455-470.
[15] RUSU R B, BLODOW N, MARTON Z C, et al. Aligning point cloud views using persistent feature histograms[C]//2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, September 22-26, 2008, Nice, France. IEEE, 2008:3384-3391.
[16] ZHENG Li, LI Zhukun. Virtual namesake point multi-source point cloud data fusion based on FPFH feature difference[J]. Sensors, 2021, 21(16):5441.
[17] RUSU R B, BLODOW N, BEETZ M. Fast point feature histograms (FPFH) for 3D registration[C]//IEEE International Conference on Robotics and Automation, May 12-17, 2009, Kobe, Japan. IEEE, 2009:1848-1853.
[18] SALTI S, TOMBARI F, DI STEFANO L. SHOT:Unique signatures of histograms for surface and texture description[J]. Computer Vision And Image Understanding, 2014, 125:251-264.
[19] CHEN Zihao, PAN Baofeng. Research on key technology of water robot avoiding collision based on improved VFH algorithm[J]. Journal of Physics, 2021, 1820(1):12064.
[20] LI Deping, LIU Ning, GUO Yulan, et al. 3D object recognition and pose estimation for random bin-picking using partition viewpoint feature histograms[J]. Pattern Recognition Letters, 2019, 128:148-154.
[21] 张铁,肖卓,邹焱飚,等. 三维点云特征的工件识别与位姿估计[J]. 机械设计与制造, 2022, 372(2):252-256. ZHANG Tie, XIAO Zhuo, ZOU Yanbiao, et al. Artifact recognition and pose estimation based on 3D point cloud feature[J]. Machine Design & Research, 2022, 372(2):252-256.
[22] LU Jun, WANG Wei, SHAO Hongxu, et al. Point cloud registration algorithm fusing of Super4PCS and ICP based on the key points[C]//2019 Chinese Control Conference (CCC), July 27-30, 2019, Guangzhou, China. IEEE, 2019:4439-4444.
[23] BESL P J, MCKAY H D. A method for registration of 3-D shapes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(2):239-256.

一种基于点云场景分割与改进配准算法的物体位姿估计方法

Object Pose Estimation Method Based on Point Cloud Scene Segmentation and Improved Registration

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	薛培林, 吴愿, 殷国栋, 刘帅鹏, 林乙蘅, 黄文涵, 张云. 基于信息融合的城市自主车辆实时目标识别[J]. 机械工程学报, 2020, 56(12): 165-173.
[2]	刘斌;上官宁;林俊义;江开勇. 脚型解剖学标志点自动标识方法[J]. , 2009, 45(11): 205-211.
[3]	杨晶东;洪炳镕;周彤;陈凤东. 移动机器人里程计非系统误差不确定性分析方法[J]. , 2008, 44(8): 7-12.