• DocumentCode
    697849
  • Title

    Tracking multiple objects using intensity-GVF snakes

  • Author

    De Vylder, Jonas ; Ochoa, Daniel ; Philips, Wilfried ; Chaerle, Laury ; Van Der Straeten, Dominique

  • Author_Institution
    Dept. of Telecommun. & Inf. Process., Ghent Univ., Ghent, Belgium
  • fYear
    2009
  • fDate
    24-28 Aug. 2009
  • Firstpage
    602
  • Lastpage
    606
  • Abstract
    Active contours or snakes are widely used for segmentation and tracking. Multiple object tracking remains a difficult task, characterised by a trade off between increasing the capturing range of edges of the object of interest, and decreasing the capturing range of other edges. We propose a new external force field which is calculated for every object independently. This new force field uses prior knowledge about the intensity of the object of interest. Using this extra information, this new force field helps in discriminating between edges of interest and other objects. For this new force field, the expected intensity of an object must be estimated. We propose a technique which calculates this estimation out of the image.
  • Keywords
    edge detection; image capture; object tracking; active contours; capturing range; external force field; intensity-GVF snakes; multiple object tracking; object edges; object intensity estimation; prior knowledge; segmentation; Active contours; Equations; Estimation; Force; Image segmentation; Shape; Vectors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Processing Conference, 2009 17th European
  • Conference_Location
    Glasgow
  • Print_ISBN
    978-161-7388-76-7
  • Type

    conf

  • Filename
    7077421