AI-Driven Diagnosis of Apple Rust and Scab Using VGG16 Framework

Abstract

Accurate detection and classification of plant diseases are essential for enabling timely intervention, reducing crop losses, and improving agricultural productivity. Apple scab and apple rust are among the most common diseases affecting apple production worldwide. This study proposes a transfer learning approach based on the Visual Geometry Group 16 (VGG16) convolutional neural network for classifying apple leaf images into three categories: healthy, apple scab, and apple rust. Apple scab is caused by the fungus Venturia inaequalis, whereas apple rust is primarily associated with Gymnosporangium juniperi-virginianae. The proposed method applies image preprocessing and a pretrained VGG16 architecture to extract robust and discriminative visual features from apple leaf images. The dataset was obtained from Kaggle and divided into training and testing sets for model development and performance evaluation. Experimental results show that the proposed model achieved a testing accuracy of 97.28%, corresponding to a misclassification rate of 2.72%. These findings demonstrate that the VGG16-based transfer learning model is effective for automated multiclass classification of apple leaf diseases and has potential for supporting practical plant disease detection applications.