On the use of agent technology in intelligent, multisensory and distributed surveillance

Abstract

This article revises the state of the art of the application of agent technology within the scope of surveillance systems. Thus, the potential of the practical use of the concepts and technologies of the agent paradigm can be identified and evaluated in this domain. Current surveillance systems are noted for using several devices, heterogeneous in many instances, distributed along the observed scenario, while incorporating a certain degree of intelligence to alert the operator proactively to what is going on in the observed scenario and prevent the operator from having to observe the monitors continuously. The basic characteristics of the agents (autonomy, reactivity, proactiveness and social ability), along with multiagent systems' characteristics (distributed data management, low coupling, robustness, communication and coordination between autonomous entities), suggest that the agency is a good choice for solving problems which appear and are dealt with in surveillance systems.

The Strategies of Optimizing Fuzzy Petri Nets By Using An Improved Genetic Algorithm

“…In many practical cases, only relying on experts' experience is very difficult to obtain those parameters accurately, sometimes even could not be obtained. As a statistical-heuristic combinatorial optimization searching method, the genetic algorithm (which is based on the biological evolution principle) is good at searching an optimal solution in a large, complex and uncertain system and searching for the best individual to meet the needs of that system from a global perspective, so it has been successfully applied to the optimization of some complicated problems in the fields such as neural networks, fuzzy processing, etc [6,7]. In this paper, the strategies based on an improved genetic algorithm are introduced to realize searching an optimal solution for the parameters of a FPN model.…”

“…In the FPN model of the Figure 3, t 1 and t 2 are the transitions of the 1st layer, t 3 , t 4 and t 5 respectively corresponds to the transition of the 2nd, 3rd and 4th layer. Because t 1 and t 3 do not lie in the same layer, but they all correspond to the same output place p 3 , in order to construct a better hierarchical model and conduce to lighting the transitions layer by layer in the process of fuzzy reasoning, we introduce a virtual place p 9 (represented by a hollow circle) and a virtual transition t 6 (represented by a small hollow rectangle) between t 1 and p 3 [1], at the same time cancel the output arc from t 1 directly pointing to p 3 , and replace it with one from t 1 pointing to the virtual place p 9 , then add an input arc from p 9 pointing to the virtual transition t 6 and an output arc from t 6 pointing to p 3 . The FPN model added in a virtual place and a virtual transition is shown in Figure 4.…”

“…The FPN model added in a virtual place and a virtual transition is shown in Figure 4. The hierarchy in the Figure is as follows: t 1 and t 2 are the transitions of the 1st layer, t 3 and t 6 are the transitions of the 2nd layer, t 4 and t 5 still corresponds to the transition of the 3rd and the 4th layer respectively. The virtual place and the virtual transition only play an Intermediate transitional role in the FPN model, adding them has no effect on the rule base system, the threshold of the virtual transition is 0, and the CF is 1.…”

“…After determining the reasoning process of the FPN model, we could commence training the above FPN model by using an improved genetic algorithm [6].…”

Context-based Information Fusion: A survey and discussion

“…In the distributed vision domain, multi-agent solutions have been proposed to exploit the coordination capability to manage multiple sensing nodes and improve the tracking results [39]. Here, the context of each vision node needs to be shared with the other ones perceiving the same scene (from differents point of view).…”