一种基于分类的复杂场景背景更新算法

提出了一种基于分类的背景更新算法。对现有模板进行改进，提出一种生长模板来对差分图像中的目标点和噪声点进行检测。生长模板根据像素点局部特征自动选择生长方向，从而对目标点和噪声点进行判断。利用基于像素的背景更新策略，实现了目标遮挡区域的背景更新。实验证明了生长模板的有效性，该算法可以在复杂场景下实现背景更新。

Background updating algorithm based on classification in complex scene

In view of the faults of traditional method, a new background updating method based on classification was proposed. A growth template that can detect the target and noise pixel was put forward. This template can choose the growth direction automatically according to the characteristics around the pixel, the template can eliminate the errors that were accumulated by the noise; background updating was implemented when many targets appeared. Comparison between the proposed method and the conventional way was made, the results of which show that the proposed method will stabilize within a very short time when quick changes happen in the scene such as the movement of the camera. This system is flexible enough to handle variations in moving objects and other arbitrary changes to the observed scene.