监控系统中的多摄像头协同算法

提出了一种监控系统中的多摄像头协同算法(PDA).该算法引入了依赖目标可见性的权函数、任务优先级、目标与摄像头之间的距离,根据这些参数来选择最合适的摄像头,将其分配给距离摄像头最近、且有最高优先级的目标.为了可靠地跟踪目标,使用了多摄像头协同系统中的数据融合方法.在分布式监控系统中验证了PDA和数据融合方法.实验结果表明,提出的PDA算法能协同多摄像头可靠地跟踪人.

An Algorithm for Cooperative Multi-camera Surveillance

An algorithm for the cooperative multi-camera surveillance system is proposed.This algorithm,which is called Priority and Distance Algorithm(PDA),introduces a weight function depending on the object visibility,the task priority and the object-to-camera distance to select the most appropriate camera,and attempts to assign the object with the highest priority to the camera with the nearest distance to the object.In order to track people reliably,a new data fusion algorithm is developed in the network of cooperative cameras,which adapts to lighting change conditions and does not need the prior knowledge of the tracking system.Both the PDA and the data fusion algorithm are implemented in the distributed surveillance system using multiple cameras to track multiple objects.Experiment results show that the system can coordinate cameras to track objects and can deal well with the occlusion.