BEAN LEAF DISEASE DETECTION USING YOLOV8 SEGMENTATION IN DEEP LEARNING

This research presents a deep learning-based approach for accurate and efficient detection of bean leaf diseases. Leveraging the YOLOv8 instance segmentation architecture, we fine-tuned the model on a curated dataset of bean leaf images. The model effectively detects and segments three primary bean leaf diseases: angular leaf spot, bean rust, and healthy leaves. Experimental results demonstrate moderate performance, achieving accuracies of 68.63% and 74.49% for original and augmented datasets, respectively. Notably, the model's confidence levels, as measured by precision, recall, F1- score, and mean average precision (MAP), improved significantly after incorporating augmented data. These results underscore the effectiveness of our approach in providing farmers with a reliable tool for monitoring crop health and implementing targeted disease management strategies.<br>