计算机应用 ›› 2019, Vol. 39 ›› Issue (11): 3355-3360.DOI: 10.11772/j.issn.1001-9081.2019040727

• 虚拟现实与多媒体计算 • 上一篇    下一篇

基于区域分割的低覆盖点云配准算法

汤慧1,2, 周明全1,3, 耿国华1   

  1. 1. 西北大学 信息科学与技术学院, 西安 710127;
    2. 西安财经大学 实验实训教学中心, 西安 710010;
    3. 北京师范大学 信息科学与技术学院, 北京 100875
  • 收稿日期:2019-04-28 修回日期:2019-06-28 发布日期:2019-08-26 出版日期:2019-11-10
  • 通讯作者: 周明全
  • 作者简介:汤慧(1982-),女,山东聊城人,实验员,博士研究生,CCF会员,主要研究方向:图形图像处理;周明全(1954-),男,陕西西安人,教授,博士生导师,硕士,主要研究方向:可视化、软件工程、中文信息处理;耿国华(1955-),女,山东莱西人,教授,博士生导师,博士,主要研究方向:图形图像处理、三维重建。
  • 基金资助:
    国家自然科学基金资助项目(61673319,61731015);青岛市自主创新重大专项(2017-4-3-2-xcl);陕西省教育厅科研计划专项(19JK0842)。

Low coverage point cloud registration algorithm based on region segmentation

TANG Hui1,2, ZHOU Mingquan1,3, GENG Guohua1   

  1. 1. College of Information Science and Technology, Northwest University, Xi'an Shaanxi 710127, China;
    2. Experimental Training Teaching Center, Xi'an University of Finance and Economics, Xi'an Shaanxi 710010, China;
    3. College of Information Science and Technology, Beijing Normal University, Beijing 100875, China
  • Received:2019-04-28 Revised:2019-06-28 Online:2019-08-26 Published:2019-11-10
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61673319, 61731015), the Major Special Project of Independent Innovation in Qingdao (2017-4-3-2-xcl), the Special Project of Scientific Research Program of Shaanxi Education Department (19JK0842).

摘要: 针对低覆盖点云配准的时间复杂度高、收敛速度缓慢以及对应点匹配易错等问题,提出一种基于区域分割的点云配准算法。首先,利用体积积分不变量计算点云上点的凹凸性,并提取凹凸特征点集;然后,采用基于混合流形谱聚类的分割算法对特征点集进行区域分割,并采用基于奇异值分解(SVD)的迭代最近点(ICP)算法对区域进行配准,从而实现点云的精确配准。实验结果表明,所提算法通过区域分割可以大幅提高点云区域的覆盖率,并且无需迭代即可计算刚体变换的最佳旋转矩阵,其配准精度比已有算法提高了10%以上,配准时间降低了20%以上。因此,所提算法是一种精度高、速度快的低覆盖点云配准算法。

关键词: 点云配准, 体积积分不变量, 区域分割, 奇异值分解, 迭代最近点

Abstract: Aiming at the problems of high time complexity, slow convergence speed and error-prone matching of low coverage point cloud registration, a point cloud registration algorithm based on region segmentation was proposed. Firstly, the volume integral invariant was used to calculate the concavity and convexity of points on the point cloud, and then the concavity and convexity feature point sets were extracted. Secondly, the regions of the feature points were partitioned by the segmentation algorithm based on the mixed manifold spectral clustering, and the regions were registered by the Iterative Closest Point (ICP) algorithm based on Singular Value Decomposition (SVD), so that the accurate registration of point clouds could be achieved. The experimental results show that the proposed algorithm can greatly improve the coverage of point clouds by region segmentation, and the optimal rotation matrix of rigid body transformation can be calculated without iteration. The algorithm has the registration accuracy increased by more than 10% and the registration time reduced by more than 20%. Therefore, the proposed algorithm can achieve fast and accurate registration of point clouds with low coverage.

Key words: point cloud registration, volume integral invariant, area segmentation, Singular Value Decomposition (SVD), Iterative Closest Point (ICP)

中图分类号: