• Acta Optica Sinica
  • Vol. 39, Issue 10, 1012001 (2019)
Mengqi Han and Wenjing Chen*
Author Affiliations
  • Department of Optoelectronic Science and Technology, College of Electronics and Information Engineering, Sichuan University, Chengdu, Sichuan 610064, China
  • show less
    DOI: 10.3788/AOS201939.1012001 Cite this Article Set citation alerts
    Mengqi Han, Wenjing Chen. Improving Measurement Accuracy of Two-Dimensional S-Transform Profilometry[J]. Acta Optica Sinica, 2019, 39(10): 1012001 Copy Citation Text show less
    Schematic of measurement geometry
    Fig. 1. Schematic of measurement geometry
    Curve-fitting processing. (a) One line of fringes; (b) upper and lower envelopes of fringes obtained by extreme points; (c) background-eliminated fringes
    Fig. 2. Curve-fitting processing. (a) One line of fringes; (b) upper and lower envelopes of fringes obtained by extreme points; (c) background-eliminated fringes
    Piecewise-mean processing. (a) One line of piecewise fringes; (b) fringes after piecewise fitting; (c) background-eliminated fringes
    Fig. 3. Piecewise-mean processing. (a) One line of piecewise fringes; (b) fringes after piecewise fitting; (c) background-eliminated fringes
    Simulated object and fringes. (a) Deformed fringes; (b) simulated object
    Fig. 4. Simulated object and fringes. (a) Deformed fringes; (b) simulated object
    Reconstructed object by global-mean processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Fig. 5. Reconstructed object by global-mean processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Reconstructed object by curve-fitting processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Fig. 6. Reconstructed object by curve-fitting processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Reconstructed object by piecewise-mean processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Fig. 7. Reconstructed object by piecewise-mean processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Reconstructed object by one-adjusting-factor processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Fig. 8. Reconstructed object by one-adjusting-factor processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Reconstructed object by two-adjusting-factor processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Fig. 9. Reconstructed object by two-adjusting-factor processing and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Reconstructed object with uncorrected phase and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Fig. 10. Reconstructed object with uncorrected phase and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Reconstructed object with second-order corrected phase and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Fig. 11. Reconstructed object with second-order corrected phase and reconstruction error. (a) Reconstructed object; (b) reconstruction error
    Experiment. (a) Reference fringes; (b) deformed fringes
    Fig. 12. Experiment. (a) Reference fringes; (b) deformed fringes
    Reconstructions by multiple methods. (a) Reconstruction by Fourier transform; (b) reconstruction by global mean; (c) reconstruction by curve fitting; (d) reconstruction by piecewise mean; (e) reconstruction by one adjusting factor; (f) reconstruction by two adjusting factors
    Fig. 13. Reconstructions by multiple methods. (a) Reconstruction by Fourier transform; (b) reconstruction by global mean; (c) reconstruction by curve fitting; (d) reconstruction by piecewise mean; (e) reconstruction by one adjusting factor; (f) reconstruction by two adjusting factors
    Errors of reconstructions by multiple methods. (a) Reconstruction error by Fourier transform; (b) reconstruction error by global mean; (c) reconstruction error by curve fitting; (d) reconstruction error by piecewise mean; (e) reconstruction error by one adjusting factor; (f) reconstruction error by two adjusting factors
    Fig. 14. Errors of reconstructions by multiple methods. (a) Reconstruction error by Fourier transform; (b) reconstruction error by global mean; (c) reconstruction error by curve fitting; (d) reconstruction error by piecewise mean; (e) reconstruction error by one adjusting factor; (f) reconstruction error by two adjusting factors
    Detailed comparisons. (a) Line 220 reconstructed surface; (b) magnification of local detail
    Fig. 15. Detailed comparisons. (a) Line 220 reconstructed surface; (b) magnification of local detail
    MethodMaximum error /mmMean of error /mmStandard deviation /mmTime /s
    Fourier transform7.361400.1781900.4951607.503960
    Global mean0.754950.1069400.17578014.120838
    Curve-fitting0.566570.0725190.13066025.083935
    Piecewise mean0.405490.0555820.09819620.676546
    One factor0.527370.0735180.12715012.993006
    Two factors0.367690.0411620.07761210.702224
    Table 1. Comparison of results by several methods
    ParameterSecond-order modifiedUnmodified
    Maximum error /mm0.5123700.799370
    Mean of error /mm0.0542010.094652
    Standard deviation /mm0.0968530.157670
    Time /s10.3439652.912564
    Table 2. Comparison of results before and after second-order phase correction
    MethodFourier transformGlobal meanCurve-fittingPiecewise meanOne factorTwo factors
    Standard deviation /mm2.68770.46280.29840.27740.30470.2745
    Table 3. Standard deviation between multiple methods and PMP
    Mengqi Han, Wenjing Chen. Improving Measurement Accuracy of Two-Dimensional S-Transform Profilometry[J]. Acta Optica Sinica, 2019, 39(10): 1012001
    Download Citation