X-LeafNet: A Modified Xception Model for Identifying Tea Leaf Diseases with Explainable Artificial Intelligence

This study presents “X-LeafNet,” a novel approach for identifying tea leaf diseases using a modified Xception model integrated with Explainable Artificial Intelligence (XAI) techniques. The system aims to classify tea leaf images into six distinct classes: Algal Spot, Brown Blight, Gray Blight, Healthy, Helopeltis, and Red Spot. Utilizing transfer learning, the modified Xception model (X-LeafNet) serves as a robust multiclass image classifier. Prior to model input, images undergo a series of preprocessing steps to enhance classification accuracy. XAI approaches, including Gradient-weighted Class Activation Mapping (Grad-CAM) and Local Interpretable Model-Agnostic Explanations (LIME), are utilized to clarify the decision-making process of the modified Xception-based convolutional neural network (CNN). These strategies enable the identification of pivotal image regions that affect the model’s predictions. Experimental results demonstrate the efficacy of X-LeafNet, achieving remarkable accuracies: 99.41% for Red Spot, 100% for Algal Spot, Brown Blight, Gray Blight, and Healthy leaves, and 99.82% for Helopeltis. With an average accuracy, precision, recall, and F1-score of 99.71%, X-LeafNet significantly outperforms existing methods, providing a reliable and interpretable solution for tea leaf disease classification.