Fire detection using the brownian correlation descriptor

Although there have been several researches for vision-based fire detection, they still have a high false positive ratio as many fire-like colored moving objects exist in the world. For this reason, we propose a new fire detection algorithm using the Brownian correlation descriptor. In this algorithm, fire candidates are detected by the chroma-intensity map based on the color property of the flame. Next, Brownian correlation descriptors are extracted from the candidates. After extracting them, the Brownian motion classifier discriminates the dynamic property between the fire and fire-like objects through multiple adjacent frames in a video. The proposed algorithm is evaluated by comparing with the recent algorithm using multiple positive and negative videos.