Multi-Agent Framework in Non-overlapping Cameras System
Chan Wang, Qiuqi Ruan · 2013
The recent researches about the surveillance of public, traffic and commercial scene is increasing the necessity to develop the intelligent visual surveillance systems, especially the multi non-overlapping cameras system for object tracking. In this paper, we use the software agent technology to build a Multi-Agent Cameras System (MACS), where each camera is represented and managed by an intelligent and collaborative agent to solve complex environment problems, such as occlusion, blind areas existing in inter-cameras, and so on. Then the camera in our system has inherited the agent's characteristics to be a camera-agent, which is not only execute the actions of object detection and tracking, but also get its knowledge by knowing about itself and the surrounding environment. What's more, the camera-agents can collaborate with each other through agents' communication mechanism to improve global cooperation of our system.