Effective S-? Background Estimation for Video Background Generation
Fan-Chei Cheng, Yu-Kumg Chen · 2008
Motion detection is a very important function in an intelligent video surveillance system. In order to detect the moving objects, it is necessary to generate the correct background image for each frame of the surveillance video. By subtracting the correct background image from the original image, the motion objects are derived in each frame of the surveillance video. Based on removing the motion objects in the Σ–Δ filter approach, a new background generation algorithm is proposed in this paper. Experiments are carried out for some surveillance videos to demonstrate the computational advantage of the proposed method.