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入。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 smart surveillance technology，more and more surveillance systems in public environments by humans have turned to be assisted by intelligent video surveillance system.To resolve the problem of background modeling initialization and updating in outdoor surveillance videos，an adaptive background modeling method is proposed.An intelligent surveillance system for transportation is also developed.Experimental results demonstrate the effectiveness and efficiency of the system.