Laser & Optoelectronics Progress, Volume. 59, Issue 5, 0512004(2022)

Defect Detection in Mirror-Like Surface Based on Phase Measuring Deflectometry

Kailong Zhang1, Li Qian1、*, and Chunlei Zhu2
Author Affiliations
  • 1School of Mechanical and Automotive Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
  • 2Suzhou Grani Vision Technology Co., Ltd, Suzhou , Jiangsu 215000, China
  • show less

    In this paper, combined with the principle of phase measuring deflectometry, the problem of defect detection is converted to the problem of extracting abrupt areas using absolute phase difference to solve the problem that traditional machine vision is difficult to detect surface defects of mirror-like surface. First, the coordinates and dimensions of the surface to be measured are obtained by gray projection method, and the area of the surface to be measured is segmented. Then, phase extraction and phase unwrapping are performed on the measured surface and the absolute phase distribution map is used as a difference with the absolute phase distribution of the reference surface to obtain an absolute phase difference map. Finally, the phase difference map is processed in frequency domain to filter the periodic components, and the Sobel filter is used to extract the phase difference map's abrupt areas to obtain surface defects. The experimental results show that the defect detection method based on phase measuring deflectometry can detect surface defects of mirror-like objects with a 0.180 mm detection accuracy.

    Tools

    Get Citation

    Copy Citation Text

    Kailong Zhang, Li Qian, Chunlei Zhu. Defect Detection in Mirror-Like Surface Based on Phase Measuring Deflectometry[J]. Laser & Optoelectronics Progress, 2022, 59(5): 0512004

    Download Citation

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

    Category: Instrumentation, Measurement and Metrology

    Received: Jul. 6, 2021

    Accepted: Sep. 16, 2021

    Published Online: Feb. 22, 2022

    The Author Email: Qian Li (ql0327@163.com)

    DOI:10.3788/LOP202259.0512004

    Topics