An Optimization Algorithm for Initial Registration of Point Clouds

被引：2
|
作者：
Shen J. [1 ]
Sun D. [1 ]
Li Y. [2 ]
Liang Z. [1 ]
机构：
[1] College of Mechanical Engineering, Shandong University of Technology, Zibo, 255049, Shandong
[2] School of Mechanical Engineering, Xi'an Jiaotong University, Xi'an
来源：
Hsi-An Chiao Tung Ta Hsueh/Journal of Xi'an Jiaotong University | 2019年 / 53卷 / 08期
关键词：
Initial registration of point clouds; Iterative closest point; Sequence images; Structure from motion;
D O I：
10.7652/xjtuxb201908022
中图分类号：
学科分类号：
摘要：
An initial registration algorithm of point clouds based on structure from motion of feature points of sequence images is proposed to solve the problem that the registration algorithms of point clouds based on corresponding points matching are difficult to adapt point clouds with large difference in initial position and their registration efficiency is low. Following the principle of perspective projection the position of a camera in the local coordinate system of a point cloud is located, and a transformation matrix that transforms the point cloud to the corresponding camera coordinate system is obtained. Then, feature points of images and their corresponding matching points are served as homonymy points, and the camera pose is globally optimized by reconstructing an image sequence. Experimental results show that the initial registration method has no strict requirements for initial position of point clouds, and obtains an approximate globally optimal initial registration for point clouds with small computational cost. When these initial registration parameters are used as initial values of an iterative closest point algorithm, the robustness of the point cloud registration is effectively improved and the computation efficiency of the iterative closest point algorithm is raised by more than 30%. © 2019, Editorial Office of Journal of Xi'an Jiaotong University. All right reserved.
引用
收藏
页码：167 / 174
页数：7
相关论文
共 20 条
  • [11] Chanki Y., Da J., A maximum feasible subsystem for globally optimal 3D point cloud registration, Sensors, 18, 2, (2018)
  • [12] Al-Manasir K., Fraser C.S., Registration of terrestrial laser scanner data using imagery, Photogrammetric Record, 21, 115, pp. 225-268, (2006)
  • [13] Wang R., Jiang G., Gao Q., Fast registration method for point clouds using the image information, Acta Geodaetica et Cartographica Sinica, 45, 1, pp. 96-102, (2016)
  • [14] Wu C., Towards linear-time incremental structure from motion, 2013 International Conference on 3D Vision, pp. 127-134, (2013)
  • [15] Lowe D.G., Distinctive image features from scale-invariant keypoints, International Journal of Computer Vision, 60, 2, pp. 91-110, (2004)
  • [16] Schonberger J.L., Frahm J.M., Structure-from-motion revisited, 29th IEEE Conference on Computer Vision and Pattern Recognition, pp. 4104-4113, (2016)
  • [17] Triggs B., Mclauchlan P.F., Hartley R.I., Et al., Bundle adjustment: a modern synthesis, International Workshop on Vision Algorithms: Theory and Practice, pp. 298-372, (2000)
  • [18] Hartley R., Zisserman A., Multiple view geometry in computer vision, Kybernetes, 30, 9-10, pp. 1865-1872, (2004)
  • [19] Snavely N., Seitz S.M., Szeliski R., Photo tourism: exploring photo collections in 3D, ACM Transactions on Graphics, 25, 3, pp. 835-846, (2006)
  • [20] Tao H., Da F., Automatic registration algorithm for the point clouds based on the normal vector, Chinese Journal of Lasers, 40, 8, pp. 184-189, (2013)
12