A Novel VGG-XGB Model for the Comprehensive Detection and Classification of Apple Leaf Diseases Using Transfer Learning

The vital role of growing apples in global farming, both for their nutritional and economic power, shows the need to tackle diseases hitting apple orchards. Using technology, especially computer-aided diagnosis systems, offers a great way to boost the spotting and sorting of diseases. This gives big gains to the farmers by upping the quality and yield of their apples. Bringing tech progress to protect apple orchards from diseases ensures a steady flow of top-notch apples for folks worldwide. This helps the Apple business grow & flourish. Pursuing technological advancements in farming, especially through computer-aided diagnosis systems, represents a significant technological breakthrough and a critical step toward ensuring the sustainability and productivity of apple farming. The goal of this research is to implement a comprehensive approach to the detection and classification of diseases in apple leaves. This proposed methodology incorporates the Wiener filter for noise removal, which boosts image clarity and adaptive thresholding for segmenting the images to find the infected areas. The VGG-16 model helps to extract the key features from the images, and XG-Boost (Extreme Gradient Boosting) is used to closely sort the different apple leaf diseases. Together, these advanced techniques have hit an impressive accuracy rate of 95.90%, showing how good this integrated approach is at boosting disease management strategies for apple orchards.