Optics and Precision Engineering, Volume. 31, Issue 4, 503(2023)

Matching point pair optimization registration method for point cloud model

Yongwei YU... Kang WANG, Liuqing DU* and Bing QU |Show fewer author(s)
Author Affiliations
  • College of Mechanical Engineering, Chongqing University of Technology, Chongqing400054, China
  • show less
    Figures & Tables(19)
    Spline with uniform noise
    Spline matching point pair distribution plot
    Curves of match point pair distance
    Improved sigmoid function curves for different
    Cross-sectional view of Top3-Bun315 registration effect
    Schematic diagram of Bunny registration result
    RMSE and registration time for different algorithms
    Registration results with different overlap rate
    Registration effect of the algorithm in this paper with different levels of noise
    Line chart of RMSE and registration time of each algorithm under different levels of noise
    Experimental environment and initial pose of the crankshaft point cloud to be registered
    Crankshaft point cloud registration results
    • Table 1. Basic Information of spline matching point pairs

      View table
      View in Article

      Table 1. Basic Information of spline matching point pairs

      Standard(2 000)With noise(2 000)With noise and after iteration(2 000)
      Count167127941
      Valid rate0.083 50.063 50.470 5
    • Table 2. Basic information of registration experiment

      View table
      View in Article

      Table 2. Basic information of registration experiment

      GroupSourceTargetξ
      1Top3Bun31545%
      2Bun270Bun31560%
      3Bun090Top375%
      4Bun000Bun04590%
    • Table 3. RMSE and registration time for different algorithms

      View table
      View in Article

      Table 3. RMSE and registration time for different algorithms

      AlgorithmRegistration data
      SourceTargetSourceTargetSourceTargetSourceBun000
      Top3Bun315Bun270Bun315Bun090Top3TargetBun045
      ICPRMSE1.78×10-40.97×10-42.83×10-41.99×10-4
      Time/s4.284.343.444.08
      Tr-ICPRMSE0.90×10-40.51×10-40.36×10-40.53×10-4
      Time/s5.134.823.814.16
      AA-ICPRMSE0.81×10-40.45×10-40.34×10-40.50×10-4
      Time/s3.153.013.133.65
      OursRMSE0.59×10-40.39×10-40.23×10-40.42×10-4
      Time/s3.473.173.293.83
    • Table 4. RMSE for different algorithms

      View table
      View in Article

      Table 4. RMSE for different algorithms

      AlgorithmNumber of source point clouds(Number of target point clouds)

      20%

      7 601(10 125)

      25%

      8 107(10 125)

      30%

      8 687(10 125)

      35%

      9 355(10 125)

      40%

      10 136(10 125)

      ICP-----
      Tr-ICP--1.72×10-41.53×10-40.95×10-4
      AA-ICP--1.53×10-41.21×10-40.84×10-4
      Ours-1.08×10-40.72×10-40.62×10-40.58×10-4
    • Table 5. RMSE and registration time of each algorithm under different levels of noise

      View table
      View in Article

      Table 5. RMSE and registration time of each algorithm under different levels of noise

      AlgorithmNumber of noise points
      N=1 000N=2 000N=3 000N=4 000N=5 000
      ICPRMSE2.78×10-42.25×10-42.46×10-41.74×10-42.59×10-4
      Time/s4.213.654.143.994.32
      Tr-ICPRMSE1.09×10-41.17×10-41.05×10-41.19×10-41.21×10-4
      Time/s5.175.315.224.985.45
      AA-ICPRMSE0.89×10-40.94×10-40.74×10-40.96×10-41.08×10-4
      Time/s3.123.273.193.203.49
      OursRMSE0.64×10-40.69×10-40.68×10-40.72×10-40.74×10-4
      Time/s3.473.573.563.323.78
    • Table 6. RMSE and time of different algorithms with gaussian noise

      View table
      View in Article

      Table 6. RMSE and time of different algorithms with gaussian noise

      DataAlgorithmRMSE/10-4Time/s

      Source:

      Bun270

      Target:

      Bun315

      ICP5.956.38
      Tr-ICP1.306.96
      AA-ICP1.385.54
      Ours0.975.87
    • Table 7. Crankshaft registration RMSE and time

      View table
      View in Article

      Table 7. Crankshaft registration RMSE and time

      DataAlgorithmRMSE/10-4Time/s

      Source:

      (143 610)

      Target:

      (129 136)

      ICP1.307.71
      Tr-ICP0.416.34
      AA-ICP0.384.29
      Ours0.275.32
    Tools

    Get Citation

    Copy Citation Text

    Yongwei YU, Kang WANG, Liuqing DU, Bing QU. Matching point pair optimization registration method for point cloud model[J]. Optics and Precision Engineering, 2023, 31(4): 503

    Download Citation

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

    Category: Information Sciences

    Received: Aug. 15, 2022

    Accepted: --

    Published Online: Mar. 7, 2023

    The Author Email: DU Liuqing (lqdu@cqut.edu.cn)

    DOI:10.37188/OPE.20233104.0503

    Topics