随着智能监控技术的快速发展, 基于数字视频的智能视频监控系统代替原来的人力监控在越来越多的公共场所获得了应用。然而, 在实际的户外监控环境下, 由于光线变化以及初始背景样本的多样性, 难以建立良好的背景样本。针对上述问题, 提出了一种自适应的背景建模方法。其中自适应指两个方面: 第一 背景的自动更新 第二, 不需要使用背景帧样本作为模型训练的输人。在此基础上, 建立了针对户外交通环境的智能车辆监控系统。实验结果表明, 上述方法在动态场景的车辆检测中取得了较好的效果。-With the rapid development of intelligent surveillance technology based intelligent video digital video surveillance system to replace the original human surveillance in public places in a growing number of applications received. However, in actual outdoor monitoring environment, due to the light changes and the diversity of the initial background samples, it is difficult to establish a good background samples. In response to these problems, proposed an adaptive background modeling method. Adaptive which refers to two aspects: first, the background is automatically updated　Second, do not need to use a background as a model for the training sample frame input. On this basis, the establishment of intelligent vehicle monitoring system for outdoor traffic environment. Experimental results show that the proposed method has achieved good results in the detection of vehicle dynamic scenes.