• DocumentCode
    2988817
  • Title

    Motion detection using periodic background estimation subtraction method

  • Author

    Shazali, S. T Syed ; Cheong, W.L. ; Mohamaddan, S. ; Kamaruddin, A. M N Abg ; Yassin, A. ; Case, K.

  • Author_Institution
    Dept. of Mech. & Manuf. Eng., Univ. Malaysia Sarawak (UNIMAS), Kota Samarahan, Malaysia
  • fYear
    2011
  • fDate
    12-13 July 2011
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    This paper proposed a motion detection system using periodic background estimation subtraction method for outdoor illumination condition using MATLAB. The proposed method is robust to illumination change effect, change in background and noise. The method basically used background subtraction. The background image is estimated at every 0.8 second when the sum of absolute difference (SAD) is less than the motion threshold. The input image is luminance normalize before background subtraction. The results were converted into binary image by autothreshold and enhanced the results with dilation and erosion. Blobs were created for each motion objects. Experiment results of using background image estimated by periodic background estimation demonstrate their robustness and effectiveness in background subtraction for real world scene.
  • Keywords
    image enhancement; image segmentation; mathematics computing; motion estimation; MATLAB; background image; binary image; illumination change effect; motion detection; motion threshold; outdoor illumination condition; periodic background estimation subtraction method; sum of absolute difference; Estimation; Image segmentation; Lighting; Motion detection; Motion segmentation; Noise; Robustness; luminance normalize; motion detection; motion threshold; periodic background estimation; sum of absolute difference;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Technology in Asia (CITA 11), 2011 7th International Conference on
  • Conference_Location
    Kuching, Sarawak
  • Print_ISBN
    978-1-61284-128-1
  • Electronic_ISBN
    978-1-61284-130-4
  • Type

    conf

  • DOI
    10.1109/CITA.2011.5999523
  • Filename
    5999523