• DocumentCode
    620042
  • Title

    High-precision maneuvering target contour tracking method based on Kalman filtering algorithm

  • Author

    Jixiang Li ; Hao Fang ; Wenjie Chen ; Mingjun Jiang

  • Author_Institution
    Sch. of Autom., Beijing Inst. of Technol., Beijing, China
  • fYear
    2013
  • fDate
    25-27 May 2013
  • Firstpage
    2041
  • Lastpage
    2045
  • Abstract
    This paper presents a high-precision maneuvering target contour tracking method based on Kalman filtering algorithm for video, against to the requirement to express the target contour accurately in engineering , and the fact that presently most of target tracking methods in video are not precise, target is easy to lose. By compensating the priori state error in update process of Kalman filtering algorithm, the method in this paper can track the target contour which occasionally moves sharply, especially it is suitable for the condition that target outline is clear against background. This method provides an approach to get target contour as well as internal information precisely, and has practical value in engineering. By the analysis of experiment, when severe maneuvering of the target happens or in the environment where there is background contour interference, the method can still track the target contour accurately, in addition it is easy to implement and has good real-time performance.
  • Keywords
    Kalman filters; target tracking; video signal processing; Kalman filtering algorithm; background contour interference; high-precision maneuvering target contour tracking method; video; Error compensation; Filtering algorithms; Interference; Kalman filters; Noise; Target tracking; Vectors; Kalman Filtering; Priori State Error Compensation; Target Contour Tracking;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control and Decision Conference (CCDC), 2013 25th Chinese
  • Conference_Location
    Guiyang
  • Print_ISBN
    978-1-4673-5533-9
  • Type

    conf

  • DOI
    10.1109/CCDC.2013.6561271
  • Filename
    6561271