针对传统单高斯背景模型(SGM)检测中背景模型不能很好地自适应背景变化等问题，提出了一种改进的单高斯背景模型检测的方法。该方法取前N 帧做均值建立初始背景模型，然后利用三帧差法计算得出背景作为本文需要处理的背景区域。同时，对帧差法获得的背景区域分区，划分出大面积静止区域、历史变化区域及该变化区域的历史轨迹区域。赋予大面积静止区和历史变化区固定更新率，同时历史变化区域的历史轨迹区域按照时间分布，给予线性衰减的更新率，在此基础上进行背景模型参数的更新，最终通过背景差分法得出运动的目标。实验表明，改进的算法背景模型的自适应性有了很大地提高，基于单高斯背景模型运动目标的检测也变得更加准确。

In order to solve the problem that background model can not be well adapt to background changes in traditional single-Gaussian background model (SGM) detection, an innovative single-Gaussian background model detection method is proposed. N frames are used to establish the initial model, and the background area is obtained by frame difference method. At the same time, the background is divided into a large area of the stationary area, historic changing area and historic area of the changing one. Then the large stationary area and historic changing area are taken with fixed update rate and the relevant track area of historic changing area is taken with linear attenuation in accordance with the time distribution. So background model parameters are updated. The moving target is obtained by using the background subtraction. According to the experiment, the self-adaptability of the proposed algorithm model is improved greatly, and the detection of moving target is more accurate based on single Gaussian background model.