Adaptive Background Updating Algorithm in Complex Scene

A common method for real-time video detection in image sequences involves ¿background subtraction¿. The numerous approaches to this problem differ in the type of background model. This paper discuss a new background updating method based on classification. A growth template that can detect the target and interference, like noise, has put forward. This template can choose the growth direction automatically according to the characteristic around the pixel, the template can eliminate the errors that accumulated by the noise; implement background updating when many targets appeared. Experiment is presented for comparison between the proposed method and the conventional way, and the results show that the proposed method will stabilize within a very short time when quick changes in the scene. This system is flexible enough to handle variations in moving objects and other arbitrary changes to the observed scene.