• Acta Photonica Sinica
  • Vol. 49, Issue 4, 0415001 (2020)
Xin-chun LI1, Zhen-yu YAN1、*, Sen LIN1、2、3, and Di JIA1
Author Affiliations
  • 1School of Electronics and Information Engineering, Liaoning Technical University, Huludao, Liaoning 125100, China
  • 2State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China
  • 3Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110016, China
  • show less
    DOI: 10.3788/gzxb20204904.0415001 Cite this Article
    Xin-chun LI, Zhen-yu YAN, Sen LIN, Di JIA. Point Cloud Registration Based on Neighborhood Characteristic Point Extraction and Matching[J]. Acta Photonica Sinica, 2020, 49(4): 0415001 Copy Citation Text show less
    Feature points extraction process
    Fig. 1. Feature points extraction process
    Normal vector direction and angle between feature point and non-feature point and neighborhood points respectively
    Fig. 2. Normal vector direction and angle between feature point and non-feature point and neighborhood points respectively
    Feature area and non-feature area
    Fig. 3. Feature area and non-feature area
    Initial matching process
    Fig. 4. Initial matching process
    Influence of threshold on initial registration error
    Fig. 5. Influence of threshold on initial registration error
    Influence of feature parameters on initial registration error
    Fig. 6. Influence of feature parameters on initial registration error
    Registration of Bunny model
    Fig. 7. Registration of Bunny model
    Registration of Dragon model
    Fig. 8. Registration of Dragon model
    Accurate registration effect of Bottle model
    Fig. 9. Accurate registration effect of Bottle model
    Noise/dBRegistration error/(×10-4mm)
    TICPMR-TICPNCC-TICPGCP-TICPNCCP-DTICP
    2564.809 007.630 856.401 624.416 816.171 0×10-2
    3063.314 006.079 554.328 432.335 808.545 7×10-2
    3565.880 005.300 552.462 223.346 605.975 1×10-2
    4067.878 004.214 904.214 516.095 903.852 5×10-2
    Table 1. Registration error of Bunny model
    Noise/dBRegistration times/s
    TICPMR-TICPNCC-TICPGCP-TICPNCCP-DTICP
    2500.2158.1973.2123.2135.11
    3000.2849.6566.2822.6430.95
    3500.3540.3746.9518.4222.17
    4000.3720.6928.2725.3318.05
    Table 2. Registration times of Bunny model
    Data lostRegistration error/(×10-4mm)
    TICPMR-TICPNCC-TICPGCP-TICPNCCP-DTICP
    10.00%77.100 003.703 503.703 274.031 702.135 6×10-2
    30.00%76.685 003.735 686.536 267.243 002.442 3×10-2
    50.00%76.652 004.085 263.268 156.548 503.160 6×10-2
    Table 3. Registration error of Dragon model
    Data lostRegistration times/s
    TICPMR-TICPNCC-TICPGCP-TICPNCCP-DTICP
    10.00%1.1989.8642.3053.4637.65
    30.00%0.7955.2332.6162.1235.49
    50.00%0.5955.5031.5548.6530.66
    Table 4. Registration times of Dragon model
    ParameterTICPMR-TICPNCC-TICPGCP-DTICPNCCP-DTICP
    Registration error/(×10-4mm)44.867 005.583 548.513 352.429 906.770 1×10-2
    Registration time/s00.6573.4731.6202.7623.71
    Table 5. Registration error and times of Bottle model
    Xin-chun LI, Zhen-yu YAN, Sen LIN, Di JIA. Point Cloud Registration Based on Neighborhood Characteristic Point Extraction and Matching[J]. Acta Photonica Sinica, 2020, 49(4): 0415001
    Download Citation