Tomato Plant Disease Detection Using YOLO11

This study focuses on the early and accurate detection of tomato plant diseases using the lightweight and efficient deep learning model YOLOv11n. Early identification of plant diseases is crucial in minimizing crop losses and ensuring sustainable agricultural practices. The research employs a structured methodology consisting of data collection, preprocessing, augmentation, and deep learning-based detection. A high-quality annotated dataset, sourced from Roboflow, was used to train the YOLOv11n model, which was fine-tuned using hyperparameter optimization to enhance detection accuracy. The model was evaluated based on precision, recall, and mean average precision (mAP), achieving an accuracy of 95 %, a recall of 94 %, and an inference time of 20 ms per image, demonstrating its suitability for real-time applications. The study highlights variability in detection performance across different disease classes, with certain disease types exhibiting higher misclassification rates, necessitating further improvements. The proposed system contributes to the advancement of smart agriculture, enabling real-time disease monitoring and precision farming applications. Future research will focus on refining the model through transfer learning and advanced augmentation techniques to enhance disease-specific detection capabilities.