Multi-Agent Framework in Visual Sensor Networks

Abstract

The recent interest in the surveillance of public, military, and commercial scenarios is increasing the need to develop and deploy intelligent and/or automated distributed visual surveillance systems. Many applications based on distributed resources use the socalled software agent technology. In this paper, a multi-agent framework is applied to coordinate videocamera-based surveillance. The ability to coordinate agents improves the global image and task distribution efficiency. In our proposal, a software agent is embedded in each camera and controls the capture parameters. Then coordination is based on the exchange of high-level messages among agents. Agents use an internal symbolic model to interpret the current situation from the messages from all other agents to improve global coordination.

A Multi-agent Architecture Based on the BDI Model for Data Fusion in Visual Sensor Networks

“…Using the proposed Multi-agent architecture, we have focused on how to fuse a set of tracks from different agents which belong to the same object. This fused information was used for two different purposes: (1) to obtain a continuity in the tracking along the field of view of the cameras involved in the distributed network, (2) to fuse only the information which provides an accurate tracking, by discarding those visual sensors which have tracking errors. The results of the first objective are shown experimentally, as the fused agent receives the tracking information of all the other agents involved in the surveillance process.…”

“…In the case of camera sensors, they cope with distributed computer vision techniques known as Visual Sensor Networks (VSNs) [2]. A distinguishing feature of visual sensors, as compared with other types of sensor such as pressure sensors, microphones or thermometers, is the considerable amount of data which they generate, which makes it essential to have local processing which delivers the information on a conceptualized level.…”

“…Using this architecture, we focus on how to fuse the tracks from different agents which pertain to the same object. This fused information is used for two different purposes: (1) to obtain continuity in the tracking along the field of view of the cameras involved in the distributed network, which provides extended coverage equivalent to that of a bigger sensor, and (2) to improve the quality of tracking using data fusion techniques, by discarding non reliable sensors. In the experiments section, results in two different scenarios with the objective of validating the previous purposes are presented.…”

Distributed Active-Camera Control Architecture Based on Multi-Agent Systems

“…So, this work proposes a MAS architecture for support advanced controls for active camera sensors in the video surveillance environment. There are some previous works done by our research group also facing the problem of still/active cameras management via MAS, like the proposed in [6,4,13]. However this previous works do not contemplate the data fusion as an integral part of the architecture itself and its management with MAS.…”

An Energy-Efficient and High-Quality Video Transmission Architecture in Wireless Video-Based Sensor Networks

“…Most of the previous works are devoted to image transmission [11-14] while the research on video transmission over WSN is still in the earlier stages. Therefore, in this article a new architecture for WVSN to transmit video streams, called Energy-efficient and high-Quality Video transmission Architecture (EQV-Architecture) is presented.…”