Efficient and Precise Disease Detection in Various Plants Using YOLOv8 for Automated Recognition

Vegetable cultivation is fundamental to global food security, yet it remains highly vulnerable to various phytopathological threats. Traditional manual disease monitoring is often reactive, labor-intensive, and subjective, frequently leading to significant yield losses due to delayed intervention. To address these challenges, this paper proposes an automated, real-time vegetable plant disease detection system leveraging the YOLOv8 (You Only Look Once, version 8) architecture. The system is designed to identify and classify multiple disease types—including fungal, bacterial, and viral infections— directly from leaf imagery. By training on a diverse dataset under varying field conditions (lighting, occlusion, and scale), the model achieves high precision and rapid inference speeds suitable for edge-device deployment. Experimental results demonstrate that the YOLOv8-based approach significantly outperforms traditional methods in both detection accuracy and processing time. This system provides a scalable solution for early disease intervention, enabling targeted treatment, reducing chemical pesticide reliance, and fostering the transition toward sustainable smart farming.