Laser & Optoelectronics Progress, Volume. 61, Issue 4, 0437004(2024)

Fast Detection Algorithm for Baggage Pallet Based on Skeleton Model

Qijun Luo, Zheng Li*, and Qingji Gao
Author Affiliations
  • Robotics Institute, Civil Aviation University of China, Tianjin 300300, China
  • show less
    Figures & Tables(15)
    Algorithm flow
    Baggage pallet and three-skeleton models. (a) Baggage pallet; (b) point cloud model; (c) border-skeleton model; (d) point-line model
    Points with two relative position distributions
    Point-line registration model
    Self-service baggage consignment experiment platform
    Results of pallet banded border point cloud and horizontal projection. (a) (b) ε=2; (c) (d) ε=4; (e) (f) ε=6
    Results of the planar point-line model registration. (a) Before registration; (b) after registration
    Process of potential energy change
    Results of border-skeleton model registration. (a) Before registration; (b) after registration
    Results of the point cloud model registration. (a) Before registration; (b) after registration
    • Table 1. Pallet samples with different integrity

      View table

      Table 1. Pallet samples with different integrity

      ParameterSample 1Sample 2Sample 3Sample 4Sample 5Sample 6
      Photo
      Point cloud
      Defect ratio /%0016.835.062.473.8
    • Table 2. Results of the different algorithm

      View table

      Table 2. Results of the different algorithm

      DataICPFPFHSHOT-ICPProposed
      Sample 1
      Sample 2
      Sample 3
      Sample 4
      Sample 5
      Sample 6
    • Table 3. Accuracy comparison of the registration algorithms

      View table

      Table 3. Accuracy comparison of the registration algorithms

      DatafRMSE /mm
      ICPSHOT-ICPFPFHProposed
      Sample 113.3311.0714.8311.87
      Sample 26.515.545.758.87
      Sample 369.1212.4261.3412.20
      Sample 431.9832.1955.0914.51
      Sample 541.9661.7686.6317.21
      Sample 631.5833.6831.6118.48
    • Table 4. Statistical experimental results

      View table

      Table 4. Statistical experimental results

      ParameterNon-pallet sampleContain pallet samplePoint cloud defective pallet sample(defective ratio)
      <10%10%‒30%30%‒50%50%‒70%>70%
      Number of samples26822212939241713
      Accuracy /%10094.110010010094.17.7
    • Table 5. Time of different registration algorithms

      View table

      Table 5. Time of different registration algorithms

      DataTime of different registration algorithms /s
      ICPSHOT-ICPFPFHProposed
      Average2.4123.9742.670.35
      Sample 11.1815.3132.960.44
      Sample 22.3628.9448.690.46
      Sample 32.8728.4634.330.38
      Sample 42.2629.5241.770.34
      Sample 52.8536.3447.270.29
      Sample 62.6325.2723.310.23
    Tools

    Get Citation

    Copy Citation Text

    Qijun Luo, Zheng Li, Qingji Gao. Fast Detection Algorithm for Baggage Pallet Based on Skeleton Model[J]. Laser & Optoelectronics Progress, 2024, 61(4): 0437004

    Download Citation

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

    Category: Digital Image Processing

    Received: Dec. 19, 2022

    Accepted: Feb. 6, 2023

    Published Online: Feb. 26, 2024

    The Author Email: Zheng Li (zli2_16@163.com)

    DOI:10.3788/LOP223358

    CSTR:32186.14.LOP223358

    Topics