Data-Efficient Pomegranate Disease Classification Using MulTaSiCl-MulTeSiSt Framework

This project focuses on the classification of pomegranate diseases using deep learning models to enhance disease detection efficiency. The dataset includes various disease categories such as "Alternaria," "Anthracnose," "Bacterial Blight," "Cercospora," and a "Healthy" class. To achieve accurate classification, we utilized deep learning models such as MobileNet, ResNet50, and VGG16, which were fine-tuned for feature extraction and classification tasks. These models were optimized to ensure high accuracy while minimizing the amount of training data required, making the solution suitable for real-world applications. A user-friendly web application, developed using Flask, HTML, CSS, and JavaScript, allows users to upload images of pomegranate leaves, which are then classified into the appropriate disease category or identified as "Healthy." This solution aims to assist farmers and agricultural professionals in early disease detection, facilitating timely intervention and improving crop health management.

Keywords: Pomegranate Disease Classification, MobileNet, ResNet50, VGG16, Deep Learning, Flask, Early Disease Detection, Agriculture, Crop Health.