Laser & Optoelectronics Progress, Volume. 56, Issue 12, 121203(2019)

Subpixel Defect Detection in Highly Reflective Workpieces Based on Zernike Moments

Tingting Liu1, Peiguang Wang1、*, and Na Zhang2
Author Affiliations
  • 1 College of Electronic Information Engineering, Hebei University, Baoding, Hebei 0 71002, China
  • 2 College of Physical Science and Technology, Hebei University, Baoding, Hebei 0 71002, China
  • show less
    Figures & Tables(10)
    Schematic of overall hardware device in detection system
    Overall flow chart of defect detection of workpieces
    Schematic of progressive image denoising algorithm based on BM3D
    Schematic of image preprocessing algorithm of highly reflective workpiece
    Variation of edge pixels of image. (a) Original image; (b) edge pixels of local gray image; (c) 3D image
    Subpixel edge extraction based on Zernike moment. (a) Original edge position; (b) edge position after rotation; (c) generated template image
    Pixel contour lines of edge positioning error
    Experimental results. (a) Original images; (b) algorithm in Ref. [21]; (c) algorithm in Ref. [22]; (d) proposed algorithm
    Comparative experimental results under different illumination. (a) Workpiece images under different illumination; (b) results of algorithm in Ref. [21]; (c) results of algorithm in Ref. [22]; (d) results of proposed algorithm
    • Table 1. Quantitative comparison of three algorithms

      View table

      Table 1. Quantitative comparison of three algorithms

      AlgorithmIn Ref. [21]In Ref. [22]Proposed
      PSNR30.2434.1842.37
      28.3331.6240.58
      30.5231.0641.28
      SSIM0.7690.8350.902
      0.7220.7850.862
      0.7340.7920.881
      Accuracy /%869296
    Tools

    Get Citation

    Copy Citation Text

    Tingting Liu, Peiguang Wang, Na Zhang. Subpixel Defect Detection in Highly Reflective Workpieces Based on Zernike Moments[J]. Laser & Optoelectronics Progress, 2019, 56(12): 121203

    Download Citation

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

    Category: Instrumentation, Measurement and Metrology

    Received: Nov. 23, 2018

    Accepted: Jan. 21, 2019

    Published Online: Jun. 13, 2019

    The Author Email: Peiguang Wang (pgwang@hbu.edu.cn)

    DOI:10.3788/LOP56.121203

    Topics