Using Python programs to process data from unmanned aerial vehicles for forest fire monitoring

Abstract

Modern challenges to environmental safety require the integration of engineering and information technologies to quickly respond to threats such as forest fires. One promising area is the use of unmanned aerial vehicles (UAVs) in combination with geographic information systems
(GIS) and data mining software. In this paper, we consider a project to create a software solution implemented in Python for monitoring and predicting the spread of forest fires. 3,000 hectares of woodlands divided into sectors were used as a conditional landfill. Each sector was analyzed based on weather data (temperature, humidity, wind speed), vegetation density, and fire hazard index calculations. The libraries Folium (for interactive mapping), Pandas (for tabular data processing) and Matplotlib (for visualization) were used for implementation. Three drones (multispectral camera, thermal imager, RGB camera), equipped with various sensors, worked on a schedule, providing continuous monitoring based on charging time. The proposed architecture of the system makes it possible to adapt it to various geographical conditions and tasks. The results are presented in the form of graphs, heat maps and tables, which makes the system convenient for
operators and analysts. The work is relevant in the field of "Engineering and Technology" and reflects the synthesis of information and engineering solutions to environmental safety problems.