Moving Target Detection in Video Monitoring System

Moving target detection is the key technique of some video monitoring system, target detection will affected the quality of surveillance directly. Some traditional methods about moving target detection, such as background differencing and coterminous frames differencing, can't solve the influence of noises coming from natural condition changing, or can't obtain integrated target contour. In this paper, algorithm model about adaptive background is proposed through improved coterminous frames differencing, then the background subtraction algorithm is used in the new background, the dynamic threshold segmentation method is used in the frame, the morphology filter is applied to reduce noises, and the contour of moving target is recognized and tracked from the image. Experimental results demonstrate that this method can reduce the influence of noise, improve the effect of moving target detection, recognize target contour clearly

[1]  E. L. Dagless,et al.  Alternative practical methods for moving object detection , 1992 .

[2]  Hironobu Fujiyoshi,et al.  Real-time human motion analysis by image skeletonization , 1998, Proceedings Fourth IEEE Workshop on Applications of Computer Vision. WACV'98 (Cat. No.98EX201).

[3]  Hironobu Fujiyoshi,et al.  Moving target classification and tracking from real-time video , 1998, Proceedings Fourth IEEE Workshop on Applications of Computer Vision. WACV'98 (Cat. No.98EX201).

[4]  Li Yu,et al.  An LMI approach to guaranteed cost control of linear uncertain time-delay systems , 1999, Autom..

[5]  Alex Pentland,et al.  Pfinder: real-time tracking of the human body , 1996, Proceedings of the Second International Conference on Automatic Face and Gesture Recognition.

[6]  Takeo Kanade,et al.  A System for Video Surveillance and Monitoring , 2000 .

[7]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .