基于球面特征的点云配准方法研究

张学昌; 唐艳梅; 梁涛

doi:10.13433/j.cnki.1003-8728.2015.1209

基于球面特征的点云配准方法研究

doi: 10.13433/j.cnki.1003-8728.2015.1209

1.
浙江大学宁波理工学院, 宁波 315100;
2.
太原科技大学机械工程学院, 太原 030024

基金项目:

国家自然科学基金项目(51075362)、浙江省自然科学基金项目(Y1100073)及宁波市产业技术创新及成果产业化重点项目(2013B10022)资助

详细信息

作者简介:
张学昌(1969-),教授,研究方向为机械工程,逆向工程,zz_zxc123@163.com

计量
- 文章访问数: 152
- HTML全文浏览量: 18
- PDF下载量: 11
- 被引次数: 0
出版历程
- 收稿日期: 2014-02-28
- 刊出日期: 2015-12-05

Research on Registration Method of Point Clouds Based on Spherical Feature

1.
School of Mechanical Engineering, Ningbo Institute of Technology, Zhejiang University, Ningbo 315100, China;
2.
School of Mechanical Engineering, Taiyuan University of Science and Technology, Taiyuan 030024, China

摘要

摘要: 针对多视点云的配准问题,提出了基于球面特征的自动配准方法。在测量的零件周围固定标准球,把零件和标准球作为一个整体进行点云测量。用球面拟合的方法求解标准球的球心坐标,并在待配准点云的球心坐标中搜索对应点,从而计算粗配准中的旋转矩阵和平移矩阵,实现点云的粗配准,采用融入球心坐标信息的改进的ICP算法(迭代最近点法)实现点云的精配准。这种方法大大缩少了粗配准中对应点的搜索范围,并实现了自动配准,提高了配准效率,改进的ICP算法增强了配准算法的鲁棒性,实例证明该方法有效。
- 自动配准算法 /
- 球面特征 /
- 点云配准 /
- 旋转矩阵 /
- 平移矩阵
Abstract: For multi-view point cloud of registration, an automatic matching algorithm based on the spherical feature was proposed. The standard balls around the measurement parts was fixed, and then the parts and standard balls as a whole was been measured. The spherical center coordinates was found via spherical fitting. Then the corresponding points in the spherical center coordinates belong to the point cloud to be registered was searched, and the corresponding points to calculate the registration of rotation matrix and translation matrix was used, those was used to point cloud of coarse registration. Finally, an improved ICP (Iterative Closest Point) algorithm was used to integrate the sphere coordinate information center to the point cloud of precise registration. The present registration method greatly reduced the coarse registration of corresponding point search, and the automatic registration was been implemented. The present algorithm also improved the registration efficiency and the improved ICP algorithm enhances the robustness of matching algorithm. The examples verifies that the method is effective.
- algorithms /
- automobiles /
- calculations /
- computational efficiency /
- efficiency

HTML全文

参考文献(17)

[1]	Salvi J, Matabosch C, Fofi D, et al. A review of recent range image registration methods with accuracy evaluation[J]. Image and Vision Computing, 2007,25(5):578-596
[2]	Hecker Y C, Bolle R M. On geometric hashing and the generalized hough transform[J]. IEEE Transactions on Systems, Man and Cybernetics, 1994,24(9):1328-1338
[3]	Wolfson H J, Rigoutsos I. Geometric hashing: an overview[J]. IEEE Computer Science & Engineering, 1997:10-21
[4]	Stockman G. Object recognition and localization via pose clustering[J]. Computer Vision, Graphics, and Image Processing, 1987,40(3):361-387
[5]	Johnson A E. Spin-images: a representation for 3-D surface matching[D]. Pittsburgh, USA: Carnegie Mellon University, 1997
[6]	Lee J D, Huang C H, Liu L C, et al. A modified soft-shape-context ICP registration system of 3-D point data[J]. IEICE Transactions on Information and Systems, 2007,E90-D(12):2087-2095
[7]	Chua C S, Jarvis R. Point signatures: a new representation for 3D object recognition[J]. International Journal of Computer Vision, 1997,25(1):63-85
[8]	Salvi J, Matabosch C, Fofi D, et al. A review of recent range image registration methods with accuracy evaluation[J]. Image and Vision Computing, 2007,25(5):578-596
[9]	Gelfand N. Feature analysis and registration of scanned surfaces[D]. Palo Alto, California: Stanford University, 2006
[10]	Besl P J, McKay N D. A method for registration of 3-D shapes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992,14(2):239-256
[11]	Rusinkiewicz S, Levoy M. Efficient variants of the ICP algorithm[C]//Proceedings of the Third International Conference on 3-D Digital Imaging and Modeling, Quebec City, Que: IEEE, 2001:145-152
[12]	Turk G, Levoy M. Zippered polygon meshes from range images[C]//Proceedings of ACM SIGGRAPH. Orlando, Florida: ACM, 1994:311-318
[13]	Masuda T, Sakaue K, Yokoya N. Registration and integration of multiple range images for 3-D model construction[C]//Proceedings of the 13th International Conference on Pattern Recognition. Vienna: IEEE, 1996:879-883
[14]	Jost T. Fast geometric matching for shape registration[D]. Neuchatel, Switzerland: University of Neuchatel, 2002
[15]	熊邦书,何明一,俞华璟.三维散乱数据的k个最近邻域快速搜索算法[J].计算机辅助设计与图形学学报,2004,16(7):909-912,917 Xiong B S, He M Y, Yu H J. Algorithm for finding k-nearest neighbors of scattered points in three dimensions[J]. Journal of Computer Aided Design & Computer Graphics, 2004,16(7):909-912,917 (in Chinese)
[16]	刘越华,廖文和,刘浩.逆向工程中散乱点云的K邻域搜索算法研究[J].机械设计与制造,2012,(3):256-258 Liu Y H, Liao W H, Liu H. Research of k-nearest neighbors search algorithm in reverse engineering[J]. Machinery Design & Manufacture, 2012,(3):256-258 (in Chinese)
[17]	陈曦.反求工程中基于点云的特征挖掘技术研究[D].杭州:浙江大学,2005 Chen X. Based on point cloud feature mining technology in reverse engineering[D]. Hangzhou: Zhejiang University, 2005 (in Chinese)