• DocumentCode
    567945
  • Title

    Emphasis of GPR signals reflected from buried pipes using statistical clustering techniques

  • Author

    Uchikado, Tomohiro ; Kobayashi, Makoto ; Nakano, Kazushi ; Shin, Seiichi

  • Author_Institution
    Univ. of Electro-Commun., Chofu, Japan
  • fYear
    2012
  • fDate
    15-17 July 2012
  • Firstpage
    410
  • Lastpage
    416
  • Abstract
    When digging paved ground, it is necessary to check in advance the location of buried pipes without damaging them. Ground-penetrating radar (GPR), which can be used for imaging without destroying underground structures, has been commonly used. However it becomes a burden for the user of GPR to visually check the location of the buried pipes from the signal received from a GPR. We have already proposed a wavelet-based method to solve this problem. We introduced a "clustering technique" for emphasizing reflected waves from pipes. Since it is based on a statistical procedure, we have the problem of showing different results for each trial. In this paper, we propose a new method for reducing the dispersion of clustering results by implementing the original method several times. This method is not for detecting the location of buried pipes, but it aims at an improvement of the precision in comparison to the original method.
  • Keywords
    ground penetrating radar; pipes; radar signal processing; statistical analysis; wavelet transforms; GPR signals; buried pipe location; buried pipes location; clustering dispersion; clustering technique; ground-penetrating radar; statistical clustering techniques; statistical procedure; underground structures; wavelet-based method; Delay; Gaussian distribution; Ground penetrating radar; Testing; Wavelet analysis; Wavelet transforms; Wavelet transform; clustering; ground-penetrating radar;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Wavelet Analysis and Pattern Recognition (ICWAPR), 2012 International Conference on
  • Conference_Location
    Xian
  • ISSN
    2158-5695
  • Print_ISBN
    978-1-4673-1534-0
  • Type

    conf

  • DOI
    10.1109/ICWAPR.2012.6294816
  • Filename
    6294816