Fast Point Cloud Splicing Algorithm Based on Weighted Neighborhood Information of Points

被引：0
|
作者：
Lu J. [1 ]
Fan Z. [1 ]
Wang W. [1 ]
机构：
[1] College of Automation, Harbin Engineering University, Harbin
来源：
Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao/Journal of Computer-Aided Design and Computer Graphics | 2019年 / 31卷 / 07期
关键词：
Feature points; Four-points congruent sets splicing algorithm; Weighted principal component analysis;
D O I：
10.3724/SP.J.1089.2019.17436
中图分类号：
O212 [数理统计];
学科分类号：
摘要：
Aiming at the problem that the traditional four-points congruent sets (4PCS) splicing algorithm is not efficient when the data volume is large, this paper proposed a point cloud fast splicing algorithm based on weighted neighborhood information of points. Firstly, a weighted principal component analysis algorithm is designed to calculate the normal vector of the point more accurately. Secondly, the distance from the point to the center of gravity of its neighborhood is used to extract the feature points. The corresponding point pairs are obtained by using neighborhood feature description based on weighted curvature estimation. The double constraint algorithm is adopted to filter the false correspondences. Finally, the extracted corresponding point pairs are used as the initial data of 4PCS splicing algorithm. The experimental results show that the weighted estimation of normal and curvature, feature points extraction and filtering methods of correspondences are stable and reliable. The splicing accuracy and efficiency of proposed point cloud splicing method are improved compared with traditional 4PCS splicing algorithm. © 2019, Beijing China Science Journal Publishing Co. Ltd. All right reserved.
引用
收藏
页码：1238 / 1246
页数：8
相关论文
共 15 条
  • [1] Chua C.S., Han F., Ho Y.K., 3D human face recognition using point signature, Proceedings of the 4th IEEE International Conference on Automatic Face and Gesture Recognition, pp. 233-238, (2000)
  • [2] Yang R.G., Allen P.K., Registering, integrating, and building CAD models from range data, Proceedings of the IEEE International Conference on Robotics and Automation, pp. 3115-3120, (1998)
  • [3] Stamos L., Leordeanu M., Automated feature-based range registration of urban scenes of large scale, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 555-561, (2003)
  • [4] Higuchi K., Hebert M., Ikeuchi K., Building 3D models from unregistered multiple range images, Systems and Computers in Janpan, 29, 6, pp. 82-90, (1998)
  • [5] Jiang J., Cheng J., Chen X.L., Registration for 3-D point cloud using angular-invariant feature, Neurocomputing, 72, 16-18, pp. 3839-3844, (2009)
  • [6] Han B., Cao J., Su Z., Automatic point clouds registration based on regions, Journal of Computer-Aided Design & Computer Graphics, 27, 2, pp. 313-319, (2015)
  • [7] Aiger D., Mitra N.J., Cohen-Or D., 4-points congruent sets for robust pairwise surface registration, ACM Transactions on Graphics, 27, 3, (2008)
  • [8] Da Silva J.P., Borges D.L., De Barros Vidal F., A dynamic approach for approximate pairwise alignment based on 4-points congruence sets of 3D points, Proceedings of the 18th IEEE International Conference on Image Processing, pp. 889-892, (2011)
  • [9] Dorai C., Weng J., Jain A.K., Optimal registration of multiple range views, Proceedings of the 12th International Conference on Pattern Recognition, pp. 569-571, (1994)
  • [10] He B.W., Lin Z.M., Li Y.F., An automatic registration algorithm for the scattered point clouds based on the curvature feature, Optics & Laser Technology, 46, pp. 53-60, (2013)
12