Automatic Recognition and Counting of Fire Blight Bacterial Disease in Apple Orchards using YOLO11

We present a novel real-time system for detecting and counting Fire Blight (FB) occurrences in apple orchards using YOLOv10 and YOLO11. Unlike previous studies focused on static images, our system operates directly on live video captured from a moving tractor. A GoPro Hero 12 Black mounted on the vehicle streams video to an onboard computer, which performs real-time processing using OpenCV and Python. With a dataset of 2,726 annotated images, YOLO11 outperformed YOLOv10 in detection accuracy. The proposed system enables efficient, on-the-go disease monitoring and supports precision agriculture through automated logging and analysis.