Real-Time Fire Detection and Monitoring using a Combination of OpenCV-based Computer Vision and Betaflight-Enabled UAV Navigation

Wildfires and outbreaks of industrial fires pose an ongoing threat to life, infrastructure and ecosystems. Traditional fire surveillance methods are often handicapped by delayed detection, restricted coverage areas and high operational costs which restrict their effectiveness in emergency response. This enterprise presents Aero Flame, an autonomous drone- based fire surveillance system which combines computer vision, real- time communication and low- cost hardware. The fire surveillance system presented uses OpenCV for the detection of fire by means of image processing, an ESP32 microcontroller for transmission of information and the BetaFlight firmware for stable flight and autonomous navigation of the drone. The drone is equipped with thermal and optical cameras to keep continuous surveillance of fire prone areas, transmitting reports of fire danger in real time to ground stations. Testing of the prototypes suggested their reliability in the detection of fire under controlled conditions with efficient handling of data and stability in flight. In terms of conventional methods, it is a system which enhances the reliability of the detection of fire, shortens the time taken in the response to fire and is economically viable an equipment suitable for use in forest, industrial and remote areas.