Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 62, Issue 4, 0415009(2025)

Robust Point Cloud Registration Method with Voxel Feature Fusion

Liangliang Qian1,2、*, Wen Nie1,2, Haosheng Zhang1,2, and Tianqiang Zhu2
Author Affiliations
  • 1School of Advanced Manufacturing, Fuzhou University, Quanzhou 362200, Fujian , China
  • 2Quanzhou Institute of Equipment Manufacturing, Haixi Institute, Chinese Academy of Sciences, Quanzhou 362200, Fujian ,China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (10)
    Equations (23)
    References (20)
    Tools
    Paper Information
    Topics
    Figures & Tables(10)
    Point cloud registration framework

    Fig. 1. Point cloud registration framework

    Download full size
    Point cloud voxel meshing

    Fig. 2. Point cloud voxel meshing

    Download full size
    SHOT feature spatial distribution map

    Fig. 3. SHOT feature spatial distribution map

    Download full size
    4PCS algorithm combining SHOT key points

    Fig. 4. 4PCS algorithm combining SHOT key points

    Download full size
    The registration effects of point clouds with different initial values and different overlap rates. (a1) (b1) (c1) (d1) (e1) Source point clouds; (a2) (b2) (c2) (d2) (e2) registration effects of Go-ICP algorithm; (a3) (b3) (c3) (d3) (e3) registration effects of Super-GICP algorithm; (a4) (b4) (c4) (d4) (e4) registration effects of proposed algorithm

    Fig. 5. The registration effects of point clouds with different initial values and different overlap rates. (a1) (b1) (c1) (d1) (e1) Source point clouds; (a2) (b2) (c2) (d2) (e2) registration effects of Go-ICP algorithm; (a3) (b3) (c3) (d3) (e3) registration effects of Super-GICP algorithm; (a4) (b4) (c4) (d4) (e4) registration effects of proposed algorithm

    Download full size
    The registration effects of point clouds with different data volumes and densities. (a1) (b1) (c1) Source point clouds; (a2) (b2) (c2) registration effects of Go-ICP algorithm; (a3) (b3) (c3) registration effects of Super-GICP algorithm; (a4) (b4) (c4) registration effects of proposed algorithm

    Fig. 6. The registration effects of point clouds with different data volumes and densities. (a1) (b1) (c1) Source point clouds; (a2) (b2) (c2) registration effects of Go-ICP algorithm; (a3) (b3) (c3) registration effects of Super-GICP algorithm; (a4) (b4) (c4) registration effects of proposed algorithm

    Download full size
    The registration effects of culvert tunnel point clouds. (a1) (b1) Source point clouds; (a2) (b2) registration effects of Go-ICP algorithm; (a3) (b3) registration effects of Super-GICP algorithm; (a4) (b4) registration effects of proposed algorithm

    Fig. 7. The registration effects of culvert tunnel point clouds. (a1) (b1) Source point clouds; (a2) (b2) registration effects of Go-ICP algorithm; (a3) (b3) registration effects of Super-GICP algorithm; (a4) (b4) registration effects of proposed algorithm

    Download full size
    Different culvert tunnel point cloud frame registration effect

    Fig. 8. Different culvert tunnel point cloud frame registration effect

    Download full size

    • Table 1. Point cloud registration results with different initial values and overlap rates

      View table

      Table 1. Point cloud registration results with different initial values and overlap rates

      Algorithmdragonbunnybuddhaarmadillostatuette
      ICP0.0012620.0464220.0235690.0745620.273600
      Go-ICP0.0011460.0362560.0225610.0532580.168265
      Super-4PCS0.0102590.0344520.0295230.0602310.223312
      3D-NDT0.0064720.0332480.0314200.0634850.198256
      Proposed algorithm0.0009230.0303490.0182470.0297610.093348

    • Table 2. Registration results of point clouds with different data volumes and densities

      View table

      Table 2. Registration results of point clouds with different data volumes and densities

      Cloud typeAlgorithmNumberIterationTime /sRMES /mmOverlap rate /%
      wolfGo-ICP340052.70.01699.1
      Super-4PCS340081.50.07292.5
      Proposed algorithm3400111.90.09482.3
      horseGo-ICP4848547452.60.12669.4
      Super-4PCS484856521.30.00891.2
      Proposed algorithm4848512033.00.00593.6
      roomGo-ICP112586992134.50.21542.6
      Super-4PCS1125867455.70.13481.4
      Proposed algorithm11258615265.40.08393.6
    Tools

    Get Citation

    Copy Citation Text

    Liangliang Qian, Wen Nie, Haosheng Zhang, Tianqiang Zhu. Robust Point Cloud Registration Method with Voxel Feature Fusion[J]. Laser & Optoelectronics Progress, 2025, 62(4): 0415009

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: Machine Vision

    Received: May. 27, 2024

    Accepted: Jul. 29, 2024

    Published Online: Feb. 26, 2025

    The Author Email:

    DOI:10.3788/LOP241371

    CSTR:32186.14.LOP241371

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology