Fire Detection using YCbCr Color Model

Abstract

The proposed method adopts rule based color model which are defined based on luminance and chrominance content present in an image. YCbCr color space effectively isolates luminance from chrominance compared to other color spaces like RGB and normalized RGB (rgb). The proposed method not only separates fire flame pixels but also isolates high temperature fire centre pixels by taking into account statistical parameters of fire image in YCbCr color space like mean and standard deviation. In this method four rules are defined to separate the true fire region. Two rules are defined for segmenting the fire region and other two rules are defined for segmenting the high temperature fire centre region. The results are obtained and tested for a 200 images and achieves 88% of higher true fire detection rate and less false detection rate. The proposed method can be used for real time forest fire detection with moving camera.