为了从复杂变化背景中鲁棒地检测、提取运动目标，提出一种基于像素层背景模型的运动目标检测算法。该算法采用快速均值漂移方法将背景帧上具有相同统计特性的像素划分为一个像素层，背景模型从而被表示为一组像素层，通过与邻域像素对应的层匹配来检测运动前景像素。实验结果表明，该方法可以实时、准确地检测运动目标，特别是在摄像机颤动等原因造成的背景时域不规则变化情况下，比经典的基于混合高斯背景模型的方法具有更好的检测效果。

A novel background model based on pixel layer for moving objects detection in a dynamic scene is presented in this paper. Fast mean shift approach is used to cluster into layers where those pixels share similar statistics. The background is then modeled as a group of pixel layers. An incoming pixel is detected as foreground if it does not adhere to these layer-models of the background. The experiments show that the proposed method performs better than the traditional MoG method under dynamic background and especially in the presence of nominal camera motion.