Deep Learning-Based Detection and Classification of Vine Diseases Using YOLO and CNN

Abstract

This paper proposes an intelligent vision-based system designed for the detection and classification of vine diseases using advanced deep learning methods. The proposed framework integrates YOLOv5 for rapid and accurate localization of diseased regions on vine leaves. The detected regions are subsequently classified using a tailored Convolutional Neural Network (CNN) architecture optimized for multi-class disease recognition. Evaluation tests show that the system achieves high detection precision and classification accuracy, demonstrating strong robustness under varying lighting conditions, leaf orientations, and background complexities. These results confirm the effectiveness of integrating YOLO-based detection with CNN-based classification, making the proposed approach a reliable and efficient tool for early diagnosis, monitoring, and management of vine diseases in precision agriculture. Moreover, the proposed system reduces manual inspection efforts and supports data-driven decision-making to enhance crop health and yield.