A computer vision based approach for detection of fire and direction control for enhanced operation of fire fighting robot

Abstract

Fire fighting is always a risk involved mission. To facilitate the fire extinguishing operations, autonomous robots are being developed in recent days. In this paper, we propose a fire fighting robot which uses a modular design concept to implement Fire detection, Path directing and Extinguishing operations. The usual trend in the previous implementation is the use of smoke detectors and physical sensors for fire detection as well as depth manipulation. Generally most sensors have low range and are sensitive to environmental changes. In this paper, we propose a computer vision based algorithm for fire detection and for directing the robot towards the detected fire, thereby overcoming the above limitations. Color segmentation is used in initial detection. Correlation is used to extract the non-static property of fire. Temperature sensor and UV-TRON sensor are used to confirm the presence of fire along with depth mapping. Finally, a water sprinkler is used to extinguish the detected fire.