A fire detection method using the flame image jumping feature

In order to solve the problem of early fire detection in the large space, a fire detection method based on flame jumping feature is proposed. A single Gaussian model of the background is established in RGB space, while the moving region is extracted by comparing the current frame with its background reference model. According to color model of the fire in RGB and HSI space, the suspicious region is obtained. Once there is suspicious region, multiple frames are captured and the jumping frequency is calculated. Experimental results indicate that the jumping feature is effective to identify the early fire.