ABSTRACT
This chapter explores the multi-agent paradigm and ontology-based knowledge representation formalisms to perform distributed data and information fusion in visual sensor networks (VSN). It describes the application of multi-agent systems in visual sensor domains. Modern VSNs involve the deployment of a number of cameras in a wide area and the management of these geographically distributed monitoring points. The agent communication ontology defines a set of concepts to describe the tracking information interchanged by the agents of the VSN. The standards for communication in multi-agent systems are defined in the Foundation of Intelligent Physical Agents specifications. VSN data processing is performed by agents at two logical levels: the tracking layer and the belief–desire–intention layer. Classical distributed visual systems work well for monitoring and surveillance tasks, but they can be improved using a multi-agent paradigm and ontology-based mechanisms. The fusion agent processes the update situation knowledge messages which are sent by sensor agents and initiates the fusion process.
