• DocumentCode
    2320806
  • Title

    An obstacle detection approach of transmission lines based on contour view synthesis

  • Author

    Yao, Gang ; Liu, Yong ; Dong, Fangmin ; Lei, Bangjun

  • Author_Institution
    Inst. of Intell. Vision & Image Inf., China Three Gorges Univ., Yichang, China
  • fYear
    2010
  • fDate
    16-20 Aug. 2010
  • Firstpage
    19
  • Lastpage
    24
  • Abstract
    An obstacle detection approach based on contour view synthesis for inspection robot on transmission line is presented in this paper. When the robot is offline, the virtual contour images for multiple view of obstacles are reconstructed, in which the contours of obstacles are extracted by means of gradient magnitude based contour extraction algorithm, and the corresponding points on the left and right image can be found by using region matching algorithm, then view synthesis technique is adopted to reconstruct multi-view of obstacles and a virtual contour image library built; When the robot is online, the obstacle can be recognized by matching the extracted contour of obstacle on transmission line and the virtual contours in the image library built beforehand. The experiment results show that ideal recognition can be achieved ground on the contour view synthesis approach for obstacle detection; the effectiveness of this method can be demonstrated by the comparing experiments as well.
  • Keywords
    gradient methods; image matching; image recognition; contour view synthesis; gradient magnitude; image library; inspection robot; obstacle detection approach; transmission lines; virtual contour images; Cameras; Image recognition; Inspection; Pixel; Power transmission lines; Robot sensing systems; Inspection robot; obstacle detection; template matching; view synthesis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Automation and Logistics (ICAL), 2010 IEEE International Conference on
  • Conference_Location
    Hong Kong and Macau
  • Print_ISBN
    978-1-4244-8375-4
  • Electronic_ISBN
    978-1-4244-8374-7
  • Type

    conf

  • DOI
    10.1109/ICAL.2010.5585378
  • Filename
    5585378