YOLO based Potato Leaf Disease Detection Method

This paper presents to accurately identify potato leaf diseases, this paper focus on a deep learning method utilizing YOLOv11. Its main objective is to differentiate between healthy leaves, late blight, and early blight. A dataset of 240 images of potato leaves 80 Images for each category was used in the study. In order to increase dataset diversity, pre-processing techniques included resizing images to 640x640 pixels and using data augmentation techniques like flipping, rotation, and brightness adjustment. With an overall accuracy of 94.1%, the YOLOv11 architecture performed better in detection than YOLOv7 and YOLOv8. Notably, YOLOv11 remained resilient under various imaging conditions and demonstrated exceptional performance in identifying subtle disease symptoms. In actual agricultural settings, this paper demonstrates the great potential of YOLOv11 for disease intervention, crop loss reduction, and the promotion of sustainable potato farming practices. The results highlight YOLOv11’s promise for real-time, in-field disease monitoring in precision agriculture, allowing for timely intervention and sustainable crop management.