中文摘要：

提出了一种基于3G网络的移动智能监控系统服务器端的设计方法。该系统主要包括视频采集压缩、视频数据发送以及危险行为告警等模块，采取基于混合高斯模型的视频运动目标检测与提取技术，以实现对于外人闯入的识别。同时为了增强系统的可靠性和实用性，利用无线上网模块通过3G网络实现了短信的实时告警功能。系统充分考虑了3G移动网络的特点和优势，有效地弥补了传统固网视频监控系统的不足和缺陷。利用计算机视觉类库OpenCV以及Visualc＋＋6．0对上述算法进行了验证，取得了很好的检测报警效果，实验结果表明了该方法的有效性。

英文摘要：

A mobile phone-based intellective surveillance system server based on 3G network is presented. The system is mainly composed of the modules for video collection and compression, video data transmission, dangerous action alarm, etc. Mixed Gaussian model is used to detect and extract moving objects, so the undefined behavior can be identified. In order to enhance the reliability and convenience of the system, a SMS alarm module is designed by using wireless intemet module through 3G network. The system takes the characteristics and advantages of the 3G network into consideration and offsets the deficiency and limitation of the traditional video surveillance system by fixed network. Then OpenCV and VC 6.0 are used to realize the algorithm, the detected result is well, actual experimental results validate the efficiency of the proposed method.