Direction-based motion detection using the MPEG2 encoder

  • Authors:

  • Josue A. Hernandez;Hiroyoshi Morita;Hector M. Perez

  • Affiliations:

  • National Polytechnic Institute, Department of Postgraduate Studies, Mexico;University of Electro-Communications, Tokyo, Japan;National Polytechnic Institute, Department of Postgraduate Studies, Mexico

  • Venue:
  • CEA'08 Proceedings of the 2nd WSEAS International Conference on Computer Engineering and Applications
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Now a day, the used of biometrics as part of the security systems have been increasing; to detect, classify and to track human motion with a high precision are part of this demand, for that reason many approaches have been proposed and, since computer vision let us to manipulated digital image sequences many of these techniques use the information content in a video stream has the main tool to do this task. In this paper we present a simple but robust technique to detect and to track human motion using MPEG2. the proposed method take advantage of the MPEG2 encoder extracting the movement vector from the P frames and filter them in order to obtain better information about real motion into the scene. Our method implies very low computational cost and low complexity compare with other techniques. Experimental results have shown good accuracy, demonstrating that the proposed technique could be reliable.