Acta Optica Sinica, Volume. 39, Issue 12, 1212005(2019)

Center Extraction of Structured Light Stripe Based on Back Propagation Neural Network

Yuehua Li, Peng Liu, Jingbo Zhou*, Youzhi Ren, and Jiangyan Jin
Author Affiliations
  • School of Mechanical Engineering, Hebei University of Science and Technology, Shijiazhuang, Hebei 050018, China
  • show less
    Figures & Tables(16)
    Selection of pixels of cross section profile
    Basic principle of center computation of each column using neural network
    Light stripes with different shapes for network training. (a) Falling stripe; (d) rising stripe; (c) horizontal stripe; (d) random stripe
    Convergence curve of root mean square error
    Gray value of stripe cross
    Histogram of center extraction error
    Center extraction results of strips with different shapes. (a) Arc stripe; (b) random stripe; (c) discontinuous stripe; (d) tooth stripe
    Sample of straight line
    Center extraction error of linear stipe for different numbers of hidden layer neurons. (a) Average value; (b) root mean square value
    Center extraction result of stripe and error comparison. (a) Center extraction result of stripe using neural network; (b) comparison of center extraction errors
    Comparison of center extraction results for different stripe qualities. (a) Original stripe; (b) under exposed stripe; (c) normal exposed stripe; (d) over exposed stripe
    • Table 1. Mean square error and error distribution 3σ value under different noises

      View table

      Table 1. Mean square error and error distribution 3σ value under different noises

      z005101520
      3σ /pixel0.05610.06270.08130.14190.1548
      Erms /pixel0.01870.02090.02720.04790.0521
    • Table 2. Center extraction error for different numbers of hidden layers

      View table

      Table 2. Center extraction error for different numbers of hidden layers

      ErrorL1L21 2L31 2
      1 2
      Erms /pixel0.15050.27800.15640.27880.15960.2722
      Eavr /pixel0.12180.22530.12610.22000.12230.2135
    • Table 3. Stripe center extraction error from network using different training samples

      View table

      Table 3. Stripe center extraction error from network using different training samples

      ErrorFig. 3(a)Fig. 3(b)Fig. 3(c)Fig. 3(d)
      Erms /pixelEavr /pixelEmd /pixel0.20930.16480.85250.15710.12090.53250.23320.18590.73600.14860.11810.4969
    • Table 4. Erms obtained by different center extraction methods for different angles between stripe and horizontal direction

      View table

      Table 4. Erms obtained by different center extraction methods for different angles between stripe and horizontal direction

      Angle /(°)Erms /pixel
      GGMStegerOur method
      00.26690.17480.1405
      200.29770.16640.1441
      400.36830.19850.1721
      600.45390.29560.2275
      801.45761.43181.2533
    • Table 5. Run time of different center extraction methods

      View table

      Table 5. Run time of different center extraction methods

      SampleRun time /s
      StegerGGMOur method
      Fig.7(a)Fig.10(a)Fig.11(a)15.294415.097215.10430.01170.01210.01250.03970.04080.0402
    Tools

    Get Citation

    Copy Citation Text

    Yuehua Li, Peng Liu, Jingbo Zhou, Youzhi Ren, Jiangyan Jin. Center Extraction of Structured Light Stripe Based on Back Propagation Neural Network[J]. Acta Optica Sinica, 2019, 39(12): 1212005

    Download Citation

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

    Category: Instrumentation, Measurement and Metrology

    Received: Jun. 27, 2019

    Accepted: Aug. 23, 2019

    Published Online: Dec. 6, 2019

    The Author Email: Jingbo Zhou (zhoujingbo@hebust.edu.cn)

    DOI:10.3788/AOS201939.1212005

    Topics