Image based smoke detection with two-dimensional local Hurst exponent

For early fire detection, smoke is considered as an important sign for it. Image based detection methods more useful than other methods which use some special sensor devices because of the cost and difficulties in setting them around target areas. When treating the image information of smoke, it is important to consider characteristics of smoke such as the semi-transparency, the non-stationary shape and so on. In this study, we use the nature of smoke and examine a smoke detection method based on a fractal nature of smoke as it is well-known that smoke is considered as fractal. We consider that the image information of smoke is a self-affine fractal. From this assumption, we can characterize the fractal nature of smoke with the exponent of fractal. We use the Hurst exponent H, which is one of the widely known exponent of fractal. We numeriacally calculate H of smoke from the relation between H and the wavelet transform of the image. So we detect the smoke area in images with H through the wavelet transform. Moreover, to obtain the accurate detection result, we use the time-accumulation technique to smoke detection results of each image. In the experiment, we examine the performance of our method based on the fractal property of smoke.