Detection of Fire Colored Pixels

Color values of moving pixels are compared with a pre-determined color distribution, which represents possible fire colors in video in RGB color space. The fire color distribution is obtained from sample images containing fire regions. Possible color values form a three dimensional point cloud in the RGB color space as shown in Fig.  (a). The cloud is represented by using a mixture of Gaussians in the RGB color space as described in [20]. A Gaussian mixture model with ten Gaussian distributions is estimated from past observations as shown in Fig.  (b).

Let $x[k,l]$ be a pixel at location $[k,l]$ with color values $[r_{[k,l]}, g_{[k,l]}, b_{[k,l]}]$. We check if the pixel lies within two standard deviations of the centers of the Gaussians to determine its nature. In other words, if a given pixel color value is inside one of the spheres shown in Fig.  (b), then it is assumed to be a fire colored pixel. We set a binary mask, called $FireColored$, which returns whether a given pixel is fire colored or not. The intersection of this mask with $Blobs$ formed in the first step is fed into the next step as a new binary mask called $Fire$.