Laser & Optoelectronics Progress, Volume. 60, Issue 6, 0612001(2023)

Digital Image Correlation Measurement Based on Pointwise Moving Least-Square Fitting

Shuang Fan1 and Yurong Guo1,2、*
Author Affiliations
  • 1College of Civil Engineering, Hunan University, Changsha 410082, Hunan, China
  • 2Key Laboratory of Building Safety and Energy Efficiency, Ministry of Education, Changsha 410082, Hunan, China
  • show less
    Figures & Tables(15)
    Schematic of DIC method principle
    Principle of pointwise moving least square method. (a) Schematic of support domain; (b) a fitting window; (c) schematic of local displacement field
    Crack treatment method. (a) Crack does not penetrate the window; (b) crack runs through the window
    DIC displacement fields in the reference image and region of interest. (a) Reference image; (b) displacement field u1; (c) displacement field u2; (d) displacement field u3
    Root mean square errors of the two methods under three kinds of deformation. (a) Displacement field; (b) strain field
    Fitting results of displacement field and strain field under the mean value of same y coordinate. (a) Displacement field; (b) strain field
    Target image and fitting results of discontinuous displacement field. (a) Target image; (b) fitting results
    sample15 test. (a) Reference image; (b) target image; (c) displacement field obtained by DIC
    Fitting results for strain field. (a) PLS-1; (b) PMLS-1; (c) PLS-3; (d) PMLS-3
    Comparison of strain field results at thin lines
    Fitting results of vertical strain field and local enlargements. (a) PLS-1; (b) PMLS-1; (c) PLS-3; (d) PMLS-3
    Comparison of vertical strain field results at thin lines
    • Table 1. Optimal window size and root mean square error of the two methods under different deformations

      View table

      Table 1. Optimal window size and root mean square error of the two methods under different deformations

      DeformationItemPLS-1PLS-2PLS-3PMLS-1PMLS-2PMLS-3
      u1DisplacementWindow size555555
      ERMS /pixel0.1550.1070.1070.1330.1070.107
      StrainWindow size5595511
      ERMS /10-33.5353.4002.7933.3603.2682.815
      u2DisplacementWindow size555555
      ERMS /pixel0.0910.0640.0640.0790.0640.064
      StrainWindow size55115513
      ERMS /10-31.5201.4841.2111.4561.4361.218
      u3DisplacementWindow size577599
      ERMS /pixel0.0440.0320.0320.0390.0320.032
      StrainWindow size55155521
      ERMS /10-30.5420.5470.4470.5400.5470.450
    • Table 2. Calculation time of the three methods

      View table

      Table 2. Calculation time of the three methods

      MethodWindow size is 5Window size is 9Window size is 13Window size is 17Window size is 21Window size is 25Window size is 29
      PLS-11.5281.7812.2913.0033.5944.1394.535
      PMLS-12.0022.6593.7085.0276.6718.60310.890
      MLS-1251.797260.140267.576272.980281.033287.159293.716
    • Table 3. Conditional numbers of matrix A

      View table

      Table 3. Conditional numbers of matrix A

      MethodSupport domain is 5Support domain is 9Support domain is 13Support domain is 17Support domain is 21Support domain is 25Support domain is 29
      MLS-18.687×1083.375×1081.766×1081.075×1087.175×1075.094×1073.781×107
      PMLS-141018.667304460.66780
      MLS-25.760×10177.267×10161.922×10167.011×10153.095×10151.551×10158.509×1014
      PMLS-213.35676.8902.860×1027.760×1021.728×1033.367×1035.966×103
      MLS-32.155×10231.450×10164.400×10169.144×10161.678×10172.177×10171.650×1017
      PMLS-353.2015.833×1024.271×1041.913×1046.374×1041.735×1054.093×105
    Tools

    Get Citation

    Copy Citation Text

    Shuang Fan, Yurong Guo. Digital Image Correlation Measurement Based on Pointwise Moving Least-Square Fitting[J]. Laser & Optoelectronics Progress, 2023, 60(6): 0612001

    Download Citation

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

    Category: Instrumentation, Measurement and Metrology

    Received: Feb. 14, 2022

    Accepted: Mar. 30, 2022

    Published Online: Mar. 16, 2023

    The Author Email: Yurong Guo (yurongguo@hnu.edu.cn)

    DOI:10.3788/LOP220724

    Topics