Acta Optica Sinica, Volume. 40, Issue 16, 1610002(2020)

Skull Point Cloud Registration Method Based on Curvature Maps

Wen Yang, Mingquan Zhou*, Bao Guo, Guohua Geng, Xiaoning Liu, and Yangyang Liu
Author Affiliations
  • College of Information Science and Technology, Northwest University, Xi′an, Shaanxi 710127, China
  • show less
    Figures & Tables(18)
    Diagram of local coordinate system and point projection. (a) Feature point p and its neighborhood point {pi} in local coordinate system op-xpypzp; (b) projection points on two-dimensional plane
    Sub-regions of regional curvature map
    Flow chart of proposed algorithm
    Iterative convergence curve
    Skull point cloud models to be registered
    Registration results of different rough registration methods. (a) PCA; (b) GA; (c) FPFH; (d) proposed method
    Two skulls to be registered
    Coarse registration results
    Fine registration results of ICP algorithm
    Fine registration results of improved ICP algorithm
    Initial position
    Coarse registration result
    Fine registration results of ICP algorithm
    Fine registration results of improved ICP algorithm
    • Table 1. Comparison of registration efficiency of different rough registration methods

      View table

      Table 1. Comparison of registration efficiency of different rough registration methods

      MethodRegistration error /mmTime-consuming /s
      PCA6.254×10-120.39
      GA5.697×10-124.96
      FPFH4.638×10-116.84
      Proposed method3.474×10-118.25
    • Table 2. Comparison of registration efficiency of skull point clouds

      View table

      Table 2. Comparison of registration efficiency of skull point clouds

      AlgorithmNumber of iterationsRegistration error /mmTime-consuming /s
      ICP463.652×10-247.83
      Improved ICP293.243×10-231.64
    • Table 3. Comparison of registration efficiency of bunny point clouds

      View table

      Table 3. Comparison of registration efficiency of bunny point clouds

      AlgorithmNumber of iterationsRegistration error /mmTime-consuming /s
      ICP328.495×10-321.63
      Improved ICP187.876×10-312.28
    • Table 4. Efficiency comparison of different registration algorithms

      View table

      Table 4. Efficiency comparison of different registration algorithms

      Point cloud modelNumber of point cloudsAlgorithmNumber of iterationsRegistration error /mmTime-consuming /s
      LO-RANSAC[26]485.689×10-247.86
      Skull to be registered210759Super-4PCS[27]354.384×10-232.75
      Go-ICP[28]534.930×10-252.59
      PICP[29]423.677×10-241.06
      Reference skull211234IRLS-ICP[30]373.256×10-236.63
      Proposed algorithm302.977×10-229.67
      LO-RANSAC[26]292.899×10-218.21
      Bunny to be registered40000Super-4PCS[27]191.556×10-213.22
      Go-ICP[28]312.017×10-220.69
      PICP[29]251.154×10-216.95
      Reference bunny40000IRLS-ICP[30]229.255×10-315.48
      Proposed algorithm177.688×10-311.94
    Tools

    Get Citation

    Copy Citation Text

    Wen Yang, Mingquan Zhou, Bao Guo, Guohua Geng, Xiaoning Liu, Yangyang Liu. Skull Point Cloud Registration Method Based on Curvature Maps[J]. Acta Optica Sinica, 2020, 40(16): 1610002

    Download Citation

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

    Category: Image Processing

    Received: Apr. 9, 2020

    Accepted: May. 18, 2020

    Published Online: Aug. 7, 2020

    The Author Email: Mingquan Zhou (nwuzmq@163.com)

    DOI:10.3788/AOS202040.1610002

    Topics