Robust Fire Detection Algorithm using Color and Shape Information

Abstract

In general, fire occurs only when 3 essentials elements come together: heat (sufficient), material (Fuel), and oxygen. Most of the available fire detection techniques, such as smoke detector, flame detector, heat detector etc, have a prob-lem with transport delay and also give a high rate of false alarm. The proposed algorithm detects fire at its early stage in both indoor and outdoor using video image processing with few false alarms. The algorithm begins by loading the selected video clip from the database developed to identify the presence or absence of fire in the video frame. In this approach, background subtraction was employed, which is comparing the current frame with the reference frame. If the result of subtraction is less than the set threshold, the difference is ignored and the next frame is taken. However, if the difference is equal to or greater than the set threshold, the frame is then subjected to colour and shape test. This is done by using combined RGB colour model and shape signature. The proposed technique was very effective in detecting a fire compared to those techniques using only motion or colour clues. Major contribution of this work is developing a database for both indoor and outdoor fire and also, an approach for detecting a fire using colour and shape signature. The proposed technique can further be improved by monitoring the fire growth.