<p>This study investigates the potential of deep learning and computer vision for automated plant disease detection and classification. This study presents a novel approach to plant disease detection and classification using a modified YOLOv11 architecture. The methodology incorporates a custom classification head, dynamic validation splitting, and comprehensive evaluation metrics to enhance performance. Leveraging two comprehensive datasets, the "Tomato Leaf Disease Dataset" and the "Plant Disease 2" dataset, we developed and evaluated a robust deep learning model. The "Tomato Leaf Disease Dataset" comprises 23,723 images across 11 classes, encompassing healthy tomato leaves and diseases like bacterial spots and early blight. The "Plant Disease 2" dataset consists of 2,904 images, focusing on diseases in beans, strawberries, and tomatoes, categorized into 12 classes. We meticulously preprocessed and augmented the datasets to enhance model generalization and performance. The deep learning model was rigorously trained and evaluated using key metrics such as accuracy, precision, recall, and F1-score. The model achieved high accuracy (97.88% and 96.90%, respectively) on both datasets. It demonstrated its effectiveness in recognizing a wide range of disease conditions with consistently high precision, recall, and F1 scores across various disease classes. This research underscores the transformative potential of deep learning in plant pathology, offering a rapid and accurate solution for disease identification. The ability to automate disease detection can significantly improve crop management strategies, leading to increased yields and reduced reliance on pesticides. This technology holds promise for enhancing agricultural sustainability and ensuring food security.</p>