" Toward a Smoke Detection System for Early Fire Alarming based on video processing and fuzzy reasoning"

چکیده مقاله

Smoke detection is  an important part of automatic fire surveillance system whose goal is to provide emergency response in the event of unexpected fire hazards. A new approach for smoke detection is proposed based on static and dynamic visual features that could be automatically extracted from video sequences which is then aggregated into a Fuzzy Inference System (FIS) and a decision making system in order to combine the features of smoke meaningfully. Experimental results on different video frames show the effectiveness of the proposed smoke detection system with respect to other state-of-the-art algorithms.