Abstract
Agricultural productivity is essential for global economic development by ensuring food security, boosting incomes and supporting employment. It enhances stability, reduces poverty and promotes sustainable growth, creating a robust foundation for overall economic progress and improved quality of life worldwide. However, crop diseases can significantly affect agricultural output and economic resources. The early detection of these diseases is essential to minimize losses and maximize production. In this study, a novel Deep Learning (DL) model called Explainable Lightweight Tomato Leaf Disease Network (XLTLDisNet) has been proposed. The proposed model has been trained and evaluated using a publicly available PlantVillage tomato leaf disease dataset containing ten classes of tomato leaf diseases including healthy images. By leveraging different data augmentation techniques, the proposed approach achieved an impressive overall accuracy of 97.24%, precision 97.20%, recall 96.70% and F1-score of 97.10%. Additionally, explainable AI techniques such as Gradient-weighted Class Activation Mapping (GRAD-CAM) and Local Interpretable Model-agnostic Explanations (LIME) have been integrated into the model to enhance the explainability and interpretability of the proposed study.