Moving object detection for video monitoring systems

In this paper, we present a real-time algorithm for moving object segmentation from image sequence of video monitoring systems. We capture structural background variation into a datasheet which represent a background model for a long image sequence. Then we differentiate moving object from the background by background subtraction. Our method can handle scenes containing moving backgrounds or illumination variations, and it achieves robust detection for different types of videos. Experimental results demonstrate that the proposed method can update the background exactly and quickly along with the variation of illumination and the shadow can be eliminated effectively.