Revolutionizing Agriculture: AI-Powered Pomegranate Fruit Disease Detection System

Pomegranate (Punica granatum) is a commercially important horticultural crop in India, widely cultivated for its nutritional and economic value. However, its yield is significantly threatened by various fruit diseases, including Alternaria fruit spot, Anthracnose, Bacterial blight, and Cercospora, which can cause substantial post-harvest and in-field losses. Traditional disease detection methods relying on manual visual inspection are time-consuming, inconsistent, and require expert knowledge, making them impractical for large-scale deployment. To address this, we propose an automated, AI-driven disease detection system based on a convolutional neural network (CNN) architecture. The model is trained on a labeled dataset comprising 5,099 pomegranate fruit images categorized into five classes—four disease types and one healthy class—sourced from a publicly available Kaggle dataset. The CNN model achieved an impressive test accuracy of approximately 92%, with precision, recall, and F1-scores close to 0.90 for most classes, demonstrating strong classification performance across diverse disease presentations. The system is deployed through a user-friendly graphical user interface (GUI) that not only visualizes the model’s prediction and confidence scores but also offers basic recommendations for disease management. By leveraging deep learning for early and accurate disease identification, this system has the potential to assist farmers, agronomists, and researchers in improving pomegranate crop monitoring, reducing diagnostic latency, and enabling targeted intervention strategies to minimize yield losses.