An intelligent real-time fire-detection method based on video processing

摘要: To achieve fully automatic surveillance of fires, an intelligent real-time fire detection method based on a 2-stage decision strategy video processing is proposed. The first stage to check if there existing by extracting fire-pixels from visual images. In color image processing, the RGB (red, green, blue) model has less computational complexity than other models and hence adopted describe pixels. function can be deduced saturation R component fire's dynamic features. second stage, number extracted pixels increasing with burning time greater someone threshold during interval, alarm given avoid leading disaster. reduce false-alarm rate, process repeated several times at adaptive thresholding way. Experimental results demonstrate that proposed very attractive for important military, social security, forest-fire alarm, commercial applications, so on.