Аналіз та цифровізація процесів моніторингу лісових екосистем із застосуванням БПЛА та інтелектуальних систем обробки даних

Abstract

This research provides a comprehensive technical evaluation of forest fire monitoring modernization through the lens of air navigation services digitization. The paper addresses the critical limitations of traditional surveillance and substantiates the transition toward Unmanned Aerial Vehicles (UAVs) operating as autonomous, intelligent nodes within a unified digital network. A central focus is placed on the proposed five-layer monitoring architecture, which integrates multi-source data collection via multispectral and RGB sensors with advanced pre-processing techniques, including orthotransformation and the generation of high-resolution Digital Surface Models (DSM). The study explores the integration of deep learning architectures, specifically the YOLO (v8/v10) family for rapid real-time object detection and U-Net or Mask R-CNN for precise canopy segmentation. By implementing Edge Computing directly on the UAV’s onboard microcomputers, the system shifts from high-bandwidth video streaming to the transmission of compressed geospatial metadata and structured coordinates. This technological leap is harmonized with the calculation of critical vegetation indices –NDVI, GNDVI, and NDRE–enabling the early identification of heat anomalies and drought-stressed zones before visible ignition occurs. Furthermore, the research emphasizes the navigational safety aspects of the system. By leveraging Remote ID protocols and ADS-B standards, the author demonstrates a viable framework for integrating autonomous environmental monitoring into the "Single Digit al Sky," ensuring seamless coordination with manned aircraft. The study employs Random Forest and CNN-based classification to evaluate post-fire ecosystem recovery within an integrated GISplatform. Ultimately, the article concludes that this digital ecosystem drastically reduces reaction times and enhances the operational resilience of emergency services, providing a scalable model for the global preservation of forest ecosystems in a technologically advanced aviation landscape.