Automated Apple Leaf Disease Classification Using Deep Convolutional Neural Networks: A Comparative Study on the Plant Village Dataset

Abstract

The early and accurate identification of plant diseases play a vital role in ensuring agricultural productivity and food security. In this study, we investigate the effectiveness of state-of-the-art convolutional neural network (CNN) architectures for the automated classification of apple leaf diseases using the Plant Village Apple dataset. Five high-performance models DenseNet-264, EfficientNet-B4, EfficientNet-B5, Inception-V3, and MobileNet-V3-Large were fine-tuned on expertly labeled images. DenseNet-264 outperformed other models, achieving an accuracy of 98.32%, precision of 97.83%, recall of 98.21%, and an F1-score of 98.02%. Inception-V3 also demonstrated competitive results, while MobileNet-V3-Large offered a compelling balance between accuracy and computational efficiency, making it suitable for deployment on mobile and edge devices. The findings highlight the capability of deep learning to deliver fast, reliable, and objective diagnostics from ordinary field images, significantly reducing the need for manual inspection. This approach holds promises for enhancing disease management, safeguarding crop yield, and supporting precision agriculture.