中文摘要：

近年来随着网络视频监控规模的不断扩大，为了提升系统的整体性能与服务质量，防止单点瓶颈的发生，需要将集群与负载均衡技术应用到视频监控系统．在借鉴常用负载均衡算法的基础上，充分考虑视频监控系统的功能与特点，将普遍使用的周期性动态反馈负载均衡算法进行了改进，引入了任务分类策略、加权最少任务策略、节点协同工作策略和节点主动通知策略等改进方案，并将改进后的算法与原算法进行对比测试．测试结果表明：改进后的动态反馈算法能够更加合理地分配负载，从而更加有效地利用集群服务器的资源，提高系统性能．

英文摘要：

In recent years, with the development of video surveillance system, in order to improve the system performance and the quality of service, the cluster technology and load balancing technology need to be applied in the video surveillance system. On the basis of common load balancing algorithms, considering the function and characteristics of video surveillance systems, an improved periodic dynamical feedback load balancing algorithm is presented. The task classification strategy, weighted least tasks strategy, nodes cooperation strategy and nodes active notification strategy are introduce and improved. Finally, the improved algorithm is compared with the original algorithm. The test results show that the improved algorithm can distribute the loads more reasonably and make use of the server resources more effectively.