矿业科学学报 (Oct 2023)

Research on the influence of light intensity on digital image correlation measurement accuracy in the testing of concrete

  • Yang Denghui,
  • Huang Chen,
  • Fan Ping,
  • Lü Jian,
  • Wei Peng,
  • Yang Liyun

DOI
https://doi.org/10.19606/j.cnki.jmst.2023.05.009
Journal volume & issue
Vol. 8, no. 5
pp. 677 – 687

Abstract

Read online

When the surface strain field of concrete structure is measured by digital image correlation(DIC)technology under complex light conditions, the decorrelation effect of the pattern will be caused by the variation in light intensity. For this problem, the optimal light intensity suitable for DIC calculating and analyzing is studied, and the influence of light intensity variation on the measurement accuracy and calculation efficiency of DIC is explored by the traditional 2D-DIC measurement system. The experimental result shows that the average gray level of 190 is the optimal light intensity. The picture is used as the reference image at the optimal light intensity. The strain error increases exponentially when the light intensity increases, and the strain error increases linearly when the light intensity decreases. The computation time and the matching rate of calculation points are insensitive to the slight range fluctuation of light intensity, while the computation time of DIC will increase 1 to 3 times, and about 25 % of calculation points will match failed because of the extreme variation of light intensity. The accuracy of experimental results is verified by Combining them with the image quality assessment index. In order to get high-quality images, the green band-pass filter imaging system is established in this paper. It can stabilize the image's brightness and contrast when the external light varies seriously. The strain error and calculation efficiency are all close to the optimal light intensity when external ambient light changes seriously. Compared with the traditional 2D-DIC measurement system, The green band-pass filter imaging system can be better applied to measure in the outdoor environment.

Keywords