• DocumentCode
    2063181
  • Title

    Power line inspection using segment measurement based on HT butterfly

  • Author

    Du, Shengzhi ; Tu, Chunling

  • Author_Institution
    Dept. Electr. & Min., Univ. of South Africa, Pretoria, South Africa
  • fYear
    2011
  • fDate
    14-16 Sept. 2011
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    This paper addresses the segment detection and measurement based on Hough Transform (HT) for unmanned aerial vehicle (UAV) power line inspection. We consider the fact that power lines usually turn their direction when they cross poles due to the gravity or designed corners, so straight line segment detection should be employed instead of simple straight line detection for these cases. Hough Transform (HT) butterflies are found suitable for image-based segments detection and measurement. The parameters of segments such as the position, the slope, the length are related to the features of the HT butterflies. Mathematical analysis is presented to demonstrate and build the relationship between the segment parameters measurement and the features of the HT butterflies. The application shows the proposed method is effective for power line inspection especially for corner detection when they cross poles.
  • Keywords
    Hough transforms; image segmentation; inspection; mathematical analysis; power cables; remotely operated vehicles; signal detection; HT butterfly; Hough transform; corner detection; image-based segments detection; mathematical analysis; power line inspection; segment measurement; straight line segment detection; unmanned aerial vehicle; Belts; Educational institutions; Image segmentation; Inspection; Object segmentation; Robots; Transforms; HT Butterfly shape; Hough Transform; Power Lines Inspection; Unmanned Aerial Vehicle (UAV);
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Processing, Communications and Computing (ICSPCC), 2011 IEEE International Conference on
  • Conference_Location
    Xi´an
  • Print_ISBN
    978-1-4577-0893-0
  • Type

    conf

  • DOI
    10.1109/ICSPCC.2011.6061553
  • Filename
    6061553