针对传统单高斯背景模型(SGM)存在的背景模型不能很好地自适应背景变化、目标检测不完整的问题，提出了一种改进的单高斯背景模型运动目标检测算法，该方法结合单高斯背景模型和mean shift原理对运动目标进行检测。取前N帧视频样本的均值作为初始背景模型，对当前帧图像进行运动目标的初检测，根据单高斯背景模型更新原理用当前帧图像对检测为背景的点进行背景模型更新，对更新后的背景模型中不属于背景点的像素点进行mean shift修正，将进行mean shift修正后得到的背景模型作为最终的背景模型，再通过背景差分法最终检测出运动目标。实验表明，改进的算法能很好地克服背景模型不能自适应背景变化的缺点，目标检测完整度比传统的单高斯模型高。

Aiming at the non-adaptive problem and incomplete detection of single Gaussian background model (SGM), an improved single Gaussian background model method for moving object detecton is proposed. This method combines SGM and the mean shift algorithm to detect moving objects. The initial background model is decided by using N frames of images, and the moving objects are detected, the pixel which belong to background points are updated according to the single Gaussian model algorithm, and the pixel which do not belong to the background points in the updated background model are corrected using the mean shift algorithm, the background model corrected by the mean shift algorithm is used as the final background model. The moving objects are detected using the background difference method. The experiments show that the improved method can overcome the non-adaptive shortcoming and have high detectivity.