A Moving Object Detection Scheme in Codestream Domain for Motion JPEG Encoded Movies

  • Masaaki Fujiyoshi
  • Yuji Tachizaki
  • Hitoshi Kiya
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6298)


This paper proposes a scheme for detecting moving objects (MOs) from a Motion JPEG (MJ) coded video recorded by a stationary camera. The proposed scheme detects MOs without decoding a compressed video, whereas the ordinary motion detecting schemes have to decompress the video. For MOs detection, the correlation based on the positive and negative sign of discrete cosine transformed (DCT) coefficients of video frames is used in the proposed scheme. A DCT sign is encoded separately from its corresponding magnitude, so the signs are directly extracted from a MJ compressed codestream, that is, no need to decompress the coded video. In the proposed scheme, MOs are detected in each 8×8-sized block of two adjacent video frames where the block is the MJ compression unit. Experimental results show that an usage of surrounding blocks in a detection decreases both false positive and false negative detections.


Motion detection Video surveillance DCT-SPC Bitstream 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Makarov, A.: Comparison of background extraction based intrusion detection algorithms. In: Proc. IEEE Int. Conf. Image Process., pp. 521–524 (1996)Google Scholar
  2. 2.
    Leykin, A., Tuceryan, M.: A vision system for automated customer tracking for marketing analysis: low level feature extraction. In: Proc. Human Activity Recognition and Modelling Workshop (2005)Google Scholar
  3. 3.
    Toyama, K., Krumm, J., Brumitt, B., Meyers, B.: Wallflower: principle and practice of background maintenance. In: Proc. IEEE Int. Conf. Comput. Vision, pp. 255–261 (1999)Google Scholar
  4. 4.
    Piccardi, M.: Background subtraction techniques: a review. In: Proc. IEEE Int. Conf. Syst., Man, Cybern., pp. 3099–3104 (2004)Google Scholar
  5. 5.
    Zen, H., Hasegawa, T., Ozawa, S.: Moving object detection from MPEG coded picture. In: Proc. IEEE Int. Conf. Image Process., vol. 4, pp. 25–29 (1999)Google Scholar
  6. 6.
    Nakajima, Y., Yoneyama, A., Yanagihara, H., Sugano, M.: Moving object detection from MPEG coded data. In: Proc. SPIE, vol. 3309, pp. 988–996 (1998)Google Scholar
  7. 7.
    Poppe, C., De Bruyne, S., Paridaens, T., Lambert, P., Van de Walle, R.: Moving object detection in the H.264/AVC compressed domain for video surveillance applications. J. Vis. Commun. Image R 20, 428–437 (2009)CrossRefGoogle Scholar
  8. 8.
    Ito, I., Kiya, H.: DCT sign-only correlation with application to image matching and the relationship with phase-only correlation. In: Proc. IEEE Int. Conf. Acoustics, Speech and Signal Process., vol. I, pp. 1237–1240 (2007)Google Scholar
  9. 9.
    Ito, I., Kiya, H.: Multiple-peak model fitting function for DCT sign phase correlation with non-integer shift precision. In: Proc. IEEE Int. Conf. Acoustics, Speech and Signal Process., pp. 449–452 (2009)Google Scholar
  10. 10.
    Kuglin, C.D., Hines, D.C.: The phase correlation image alignment method. In: Proc. IEEE Int. Conf. Cybern. Soc., pp. 163–165 (1975)Google Scholar
  11. 11.
    Information technology — Digital compression and coding of continuous-tone still images — Requirements and guidelines, ISO/IEC Int. Std. 10918-1 (1994)Google Scholar
  12. 12.
    Video Surveillance Online Repository (ViSOR),
  13. 13.
    CIPR Sequences, Center for Image Processing Research (CIPR) at Rensselaer Polytechnic Institute,
  14. 14.
    Datasets for 2001 IEEE Int. Workshop Performance Evaluation of Tracking and Surveillance (PETS) (2001),

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Masaaki Fujiyoshi
    • 1
  • Yuji Tachizaki
    • 1
  • Hitoshi Kiya
    • 1
  1. 1.Dept. Inform. & Commun. Systems Eng.Tokyo Metropolitan UniversityTokyoJapan

Personalised recommendations