作者：庄仁诚1,2, 陈鹏1, 傅瑶1, 黄运华1
作者单位：1. 西南交通大学机械工程学院，四川 成都 611756;
2. 哈尔滨工业大学机电工程学院，黑龙江 哈尔滨 150080
关键词：列车车轮;三维检测;结构光;点云处理
摘要：
列车轮对作为转向架的关键零部件，其检测手段仍以人工检测为主，现有的自动检测方案，大多针对车轮某一断面的参数尺寸进行测量，难以真实反映车轮轮缘踏面的损伤情况。为此，该文提出一种列车车轮三维结构光检测中的点云处理方案。首先，利用三维结构光测量仪器采集列车车轮的三维点云数据；其次，根据列车车轮三维点云的特点，确定包括离群点去除、点云配准、点云平滑处理以及孔洞修补在内的点云处理方案，并对各处理步骤的最优参数进行分析；最后，利用贪婪投影三角化算法，进行列车车轮三维点云数据的曲面重建，使用拉普拉斯平滑算法对重建后的曲面进行平滑处理。结果表明，该文所提出的列车车轮点云处理方案能够实现对三维点云数据的处理，最终得到的列车车轮的三维曲面模型与基准模型的标准偏差为1.768 mm，实现对于列车车轮的三维检测。
Research on point cloud processing in train wheels three-dimensional structured light inspection
ZHUANG Rencheng1,2, CHEN Peng1, FU Yao1, HUANG Yunhua1
1. College of Mechanical Engineering, Southwest Jiaotong University, Chengdu 611756, China;
2. School of Mechatronics Engineering, Harbin Institute of Technology, Harbin 150080, China
Abstract: As an important component of the bogie, the main inspection method of the train wheelset is still the manual inspection. Current automatic inspection methods mostly measure the parameters of a certain section of the wheel, which cannot truly reflect the damage of the wheel flange and tread. For this reason, this paper proposes a point cloud processing scheme in the three-dimensional (3D) structured light inspection of train wheels. First, the 3D point cloud data of the train wheels were acquired using a 3D structured light measuring instrument. Secondly, based on the characteristics of the 3D point cloud of the train wheels, the point cloud processing scheme was studied including outlier removal, point cloud registration, point cloud smoothing and hole repair. At the same time, the optimal parameters of each processing step were analyzed. Finally, with the greedy projection triangulation algorithm, the surface reconstruction of the 3D point cloud data of the train wheels was implemented, and the reconstructed surface was smoothed by the Laplacian smoothing algorithm. The results show that the train wheel point cloud processing scheme proposed in this paper can realize the processing of the 3D point cloud data. The standard deviation of the final 3D surface model of the train wheels and the reference model is 1.768 mm, which realizes the 3D inspection of the train wheels.
Keywords: train wheels;three-dimensional inspection;structured light;point cloud processing
2021, 47(2):19-25 收稿日期: 2020-06-30;收到修改稿日期: 2020-08-15
基金项目:
作者简介: 庄仁诚(1996-),男,山东青岛市人,博士研究生,主要从事光电检测及机器视觉研究
参考文献
[1] 国家铁路局. 铁路客车轮轴组装检修及管理规则[M]. 北京: 中国铁道出版社, 2013.
[2] 冯其波, 崔建英, 陈士谦. 基于平行四边形机构的车轮几何参数自动测量方法的研究[J]. 机械工程学报, 2004(9): 190-194
[3] 左建勇, 周文祥, 曾京, 等. 应用激光传感器测量轮辋尺寸的试验研究[J]. 铁道车辆, 2002(2): 11-13
[4] BRIZUELA J, FRITSCH C, IBÁÑEZ A. Railway wheel-flat detection and measurement by ultrasound[J]. Transportation Research Part C: Emerging Technologies, 2011, 19(6): 975-984
[5] 高岩. 轮对几何尺寸自动与动态在线测量方法的研究[D]. 北京: 北京交通大学, 2014.
[6] 程宏钊. 三维检测技术在列车车轮检测与维修中的应用研究[D]. 成都: 西南交通大学, 2013.
[7] CHEN Y, MEDIONI G. Object modelling by registration of multiple range images[J]. Image and Vision Computing, 1992, 10(3): 145-155
[8] 徐思雨, 祝继华, 田智强, 等. 逐步求精的多视角点云配准方法[J]. 自动化学报, 2019, 45(8): 1486-1494
[9] 李彩林, 郭宝云, 季铮. 多视角三维激光点云全局优化整体配准算法[J]. 测绘学报, 2015, 44(2): 183
[10] LI Y, LIU Y, SUN R, et al. Multi-view point cloud registration with adaptive convergence threshold and its application in 3D model retrieval[J]. Multimedia Tools and Applications, 2019: 1-18
[11] ZHU J, GUO R, LI Z, et al. Registration of multi-view point sets under the perspective of expectation-maximization[J]. IEEE Transactions on Image Processing, 2020, 29: 9176-9189
[12] ZHU J, ZHU L, LI Z, et al. Automatic multi-view registration of unordered range scans without feature extraction[J]. Neurocomputing, 2016, 171: 1444-1453
[13] 张爽. 列车轮对几何参数在线检测关键技术研究[D]. 长春: 吉林大学, 2017.
[14] 程俊廷, 赵灿, 王从军, 等. 基于参考点和 ICP 算法的点云数据重定位研究[J]. 计算机测量与控制, 2006, 14(9): 1222-1224
[15] 何东健, 邵小宁, 王丹, 等. Kinect获取植物三维点云数据的去噪方法[J]. 农业机械学报, 2016, 47(1): 331-336
[16] 袁建英, 刘先勇, 刘伟, 等. 改进 ICP 算法实现多视点云精确配准研究[J]. 传感器与微系统, 2008, 27(5): 27-30
[17] 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
[18] 杨英保, 刘先勇, 杨俊平. 基于标志点的多视点云优化拼接方法[J]. 现代电子技术, 2011, 34(10): 20-23
[19] 程玉民. 移动最小二乘法研究进展与述评[J]. 计算机辅助工程, 2009, 18(2): 5-11
[20] ALEXA M, BEHR J, COHEN-OR D, et al. Computing and rendering point set surfaces[J]. IEEE Transactions on Visualization and Computer Graphics, 2003, 9(1): 3-15
[21] 郭浩. 点云库PCL从入门到精通[M]. 北京: 机械工业出版社, 2018.
[22] 吴旭, 卢凌雯, 梁栋栋, 等. 基于点云数据的曲面重建算法比较研究[J]. 安徽师范大学学报(自然科学版), 2019, 42(1): 46-50
[23] COHEN-STEINER D, DA F. A greedy delaunay-based surface reconstruction algorithm[J]. The Visual Computer, 2004, 20(1): 4-16
[24] 王卉, 黄玉清. 基于拉普拉斯算子的三角网格模型的平滑与压缩算法[J]. 计算机系统应用, 2015, 24(12): 191-195
