• DocumentCode
    2107336
  • Title

    Scene abrupt change detection

  • Author

    Wang, Xinying ; Weng, Zhengke

  • Author_Institution
    Dept. of Electr., Xinjiang Univ., Urumqi, China
  • Volume
    2
  • fYear
    2000
  • fDate
    2000
  • Firstpage
    880
  • Abstract
    This paper introduces a new method for video scene change detection. The proposed method utilizes the luminance histogram twice difference in order to determine the dynamic threshold needed to evaluate the break. The adaptive determination of the threshold, minimizes the number of incorrect decisions leading to a robust and accurate determination of the actual scene break. The method is simple, computationally attractive and capable of detecting changes in a variety of visual inputs. Experimental results indicate that the new method constantly outperforms existing techniques that are based on static thresholds
  • Keywords
    adaptive signal detection; brightness; image colour analysis; image segmentation; image sequences; video signal processing; adaptive change detection; adaptive thresholds; color space; dynamic threshold; experimental results; luminance histogram; scene abrupt change detection; scene break determination; static thresholds; temporal segmentation; twice difference algorithm; video scene change detection; video sequence; visual inputs; Cameras; Graphics; Gunshot detection systems; Histograms; Image segmentation; Indexing; Layout; Robustness; Streaming media; Video sequences;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electrical and Computer Engineering, 2000 Canadian Conference on
  • Conference_Location
    Halifax, NS
  • ISSN
    0840-7789
  • Print_ISBN
    0-7803-5957-7
  • Type

    conf

  • DOI
    10.1109/CCECE.2000.849592
  • Filename
    849592