• DocumentCode
    1044130
  • Title

    Accurate extraction of infrared target based on graph cut

  • Author

    Yu, S.Y. ; Zhang, Y. ; Mao, X.N. ; Yang, J.

  • Author_Institution
    Shanghai Jiaotong Univ., Shanghai
  • Volume
    44
  • Issue
    2
  • fYear
    2008
  • Firstpage
    100
  • Lastpage
    101
  • Abstract
    A novel coarse-to-fine approach to extract an infrared target is proposed. First, a confined region which contains the entire object is obtained based on the weighted information entropy feature. Secondly, according to the confined region, the energy function is constructed by combining boundary and region properties of the target region. Finally, the target is accurately extracted by energy minimisation as a graph cut problem. The experimental results show that the proposed method is robust and effective.
  • Keywords
    feature extraction; graph theory; infrared imaging; object detection; coarse-to-fine approach; energy function; graph cut; infrared target extraction; weighted information entropy feature;
  • fLanguage
    English
  • Journal_Title
    Electronics Letters
  • Publisher
    iet
  • ISSN
    0013-5194
  • Type

    jour

  • DOI
    10.1049/el:20083166
  • Filename
    4436138