Recent Advancements in Deep Learning Models for Tomato Crop Disease Detection

Recent advances in deep learning have enabled effective automated crop disease detection, which is vital for improving agricultural productivity and food security. State-of-the-art models such as Convolutional Neural Networks (CNNs) and object detection frameworks, including YOLOv8 and YOLOv11, demonstrate strong performance under complex environmental conditions. This work investigates the YOLOv8 architecture for crop disease detection, focusing on dataset preparation, image preprocessing, evaluation metrics, and deployment challenges. Transfer learning and attention mechanisms are employed to enhance performance on limited and imbalanced datasets. Key challenges such as illumination variation, occlusion, and model generalization are addressed. Experimental results show that the proposed approach achieves an average accuracy of YOLOv8 Model approximately 80%, on tomato leaf disease of plant village dataset, and YOLOv11 Model approximately 94%, on tomato leaf disease indicating its suitability for real-time precision agriculture applications.