engleski

False Alarm Reduction in Forest fire Video Monitoring System

Forest fires represent a constant threat to ecological systems, infrastructure and human lives. Great efforts are therefore made to achieve early forest fire detection, which is traditionally based on human surveillance. A rather more advanced approach to human forest fire surveillance is installation of remotely controlled video cameras sensitive in visible spectra for automatic surveillance and early forest fire detection. In Croatia exists one such system called Intelligent Forest Fire Monitoring System (iForestFire® ) developed at Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture University of Split. The system is theoretically based on the observer theory and implemented using agent technology. By now it was successfully implemented in various Croatian regions, National and Nature Parks. This paper describes one particular problem encountered in forest fire monitoring systems based on cameras sensitive in visible spectra – the problem of false alarm reduction. False alarms have been analyzed in spirit of the observer theory taking into account experimental experiences. Various forest fire false alarm causes have been identified and few methods for their reduction proposed based on digital image processing and analysis and/or fusion of information from various sources, primarily fusion of local meteorological data and information extracted from video streams.

forest fire detection; false alarms reduction

nije evidentirano