Journal of The Japanese Society for Non-Destructive Inspection
Print ISSN : 0367-5866
Volume 56, Issue 7
Displaying 1-2 of 2 articles from this issue
  • Masaki WAKABAYASHI, Shogo TANAKA
    2007 Volume 56 Issue 7 Pages 364-370
    Published: 2007
    Released on J-STAGE: July 20, 2007
    JOURNAL FREE ACCESS
    This paper analyzes the electromagnetic wave propagation paths of a radar in reinforced concrete using Snell's law. Using the derived paths, the depth and diameter of the steel bars are measured by making a pattern matching between the modeled received wave signal and the actual received wave signal of the radar. The relative dielectric constant of the concrete is also measured together with the depth and the diameter under the assumption that it is almost the same in depth direction, because the propagation speed is a function of the relative dielectric constant.Lastly, the effectiveness of the method is demonstrated by experiments; especially, the depth is measured with high accuracy. For the diameter, the error of one or two standard units (the same order of error as found with the electromagnetic induction method) is obtained and thus an additional device is suggested to improve the measurement accuracy.
    Download PDF (658K)
  • Yusuke FUJITA, Yoshihiro MITANI, Yoshihiko HAMAMOTO
    2007 Volume 56 Issue 7 Pages 371-377
    Published: 2007
    Released on J-STAGE: July 20, 2007
    JOURNAL FREE ACCESS
    This paper proposes a crack extraction method using image processing techniques. Noise included in an image, such as irregular shading and blemishes, makes crack extraction difficult. We propose a new method to subtract from an original concrete surface image the smoothed one with a median filter. The difference between the two images retains only lines and points like a crack. This operation removes irregular shading and blemishes. We also propose a method to emphasize line structures and a technique to extend broken lines with a Hessian matrix. A comparative study of 60 actual images shows that this technique detects cracks clearly.
    Download PDF (1148K)
feedback
Top