Improving Fire Detection Reliability by a Combination of Videoanalytics

In this paper we propose a novel method for detecting fires in both indoor and outdoor environments. The videos acquired by traditional surveillance cameras are analyzed and different typologies of information, respectively based on color and movement, are combined into a multi expert system in order to increase the overall reliability of the approach, making it possible its usage in real applications. The proposed algorithm has been tested on a very large dataset acquired in real environments and downloaded on the web. The obtained results confirm a consistent reduction in the number of false positive detected by the system, without paying in terms of accuracy.