FOREST FIRE MONITORING AND INTELLIGENT EARLY WARNING BASED ON MULTI-SOURCE PERCEPTION USING FIXED-WING UAVS

Abstract

Forest fires are characterized by sudden occurrence, rapid spread, and wide-ranging destruction, posing serious threats to forest resource protection and ecological security management. To address the limitations of conventional monitoring methods, such as manual patrols, fixed video surveillance, and satellite remote sensing, in terms of insufficient coverage, limited real-time capability, and low early-warning efficiency in long-distance and large-area forest fire monitoring, this study proposes a forest fire monitoring and intelligent early warning framework based on multi-source perception using fixed-wing UAVs. In the proposed framework, a fixed-wing UAV serves as the airborne patrol platform and is equipped with sensing devices such as RGB cameras, thermal infrared cameras, and multispectral sensors. The UAV conducts large-area patrols over the target forest region along predefined flight routes and collects fire-related information from multiple dimensions, including smoke, flames, thermal anomalies, and vegetation conditions. Through integrated analysis of multi-source information, the system can preliminarily identify suspected fire points, potential fire-risk areas, and fire warning levels, and transmit the monitoring results to the ground control platform to support early fire detection and emergency response. The study indicates that the combination of the long-endurance and large-coverage capability of fixed-wing UAVs with multi-source perception technology can improve the efficiency, reliability, and intelligence of forest fire monitoring, showing practical value for forest fire prevention and ecological resource protection.