Patronus: Fire Detection and Extinguisher System using Image Processing

In the modern world, extinguishers must be installed in every building. However, when a fire breaks out, there is uncertainty and confusion on whether to evacuate the area or activate the extinguisher and put out the flames. Early detection of fire is necessary to prevent the further loss of life and property. The existing systems for detecting and extinguishing a fire are the smoke sensor and sprinkler systems, which detect fire from smoke and are programmed to activate when the temperature reaches a predetermined threshold. Overall, with this type of framework, there are several disadvantages such as bogus caution, space inclusion, signal transmission, and alarm postponement. In this paper, upon conducting an extensive comparative analysis on the existing literature, a model for fire detection and extinguishing using image processing and machine learning is proposed. This technology can be used to monitor high-risk zones where there is always the possibility of a fire. The objective of this work is to provide a reliable, secure, and shrewd framework to lessen constraints and deficiencies like false alarms, which cause fear among people with the use of various advances, as well as to make the area safe from the perilous fire.