Mapping of Agricultural Land to Detect Plant Disease Pests Based on Drone

The use of drone and deep learning technology has provided a means to observe agricultural land with more efficiency and accuracy. However, manually identifying crop pests and diseases remains a slow process with the potential for human error. This study will analyze the potential of the YOLO algorithm for detection and classification of plant infections in aerial views from drone satellite images, and proposes a drone-based agricultural land mapping system for automated detection of plant pests and diseases. The results showed that the system created an interactive map of plot location and type of infection with very high accuracy, where the best results were achieved with the YOLOv8m configuration with an F1 Score of 98.3 % allowing farmers to engage in timely and informed decisions.