• DocumentCode
    2035350
  • Title

    Techniques for estimation of biological contamination on insulators using image analysis

  • Author

    Dernfalk, A.D. ; Gubanski, Stanislaw

  • Author_Institution
    Dept. of Electr. Power Eng., Chalmers Univ. of Technol., Sweden
  • fYear
    2004
  • fDate
    17-20 Oct. 2004
  • Firstpage
    659
  • Lastpage
    662
  • Abstract
    A new technique for measurement and characterization of biological contamination on composite insulators has been developed and investigated. The technique, based on photography and digital image analysis, has been applied to insulators partly covered by fungal growth. Different methods for estimating measures characterizing growth patterns, like covered area and remaining creepage distance, have been explored. It is shown that the developed technique for automatic insulator recognition and estimation of its position allows for non-supervised measurements to be performed outdoors. Finally, results from flashover voltage measurements on insulators with fungal growth are reported.
  • Keywords
    composite insulators; edge detection; flashover; image classification; image segmentation; insulator contamination; automatic insulator recognition; composite insulators; contamination characterization; digital image analysis; edge detection; flashover voltage measurements; fungal covered area; fungal growth patterns; image classification; image segmentation; insulator biological contamination estimation; nonsupervised measurements; photography; remaining creepage distance; Area measurement; Contamination; Digital images; Flashover; Image analysis; Insulation; Performance evaluation; Photography; Pollution measurement; Position measurement;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electrical Insulation and Dielectric Phenomena, 2004. CEIDP '04. 2004 Annual Report Conference on
  • Print_ISBN
    0-7803-8584-5
  • Type

    conf

  • DOI
    10.1109/CEIDP.2004.1364336
  • Filename
    1364336