Fire detection is an important topic in the field of security, researching intelligent and robust fire detection method and system have a very important theoretical and application value. This paper systematically describes the basic principle of multi-sensor information fusion technology, the principles of fire detection and the basic fire detection algorithm, and then presents a fuzzy neural network fire detection method based on T-S model in feature level fusion. The simulation results show that this method has advantages of fast identified speed and high accuracy. In order to improve robust capability of the system and based on the characteristics of fire, this paper introduces the characteristics of smoke-signal duration, and uses fuzzy logic in decision-making level, and then makes a simulation. Experimental results show that this method has good robustness and effectiveness. We put it into application of intelligent building monitoring system, and its usefulness is proved.
|