Amplifying Apple Mosaic Illness Detection: Combining CNN and Random Forest Models

Global apple harvests are seriously threatened by Apple Mosaic Disease (AMD), which calls for accurate identification and scalable control measures. This study uses Deep Learning (DL) Convolutional Neural Networks (CNN) and Random Forest (RF) models to investigate AMD categorization across four severity levels. The study carefully chooses a broad leaf dataset that includes both healthy and diseased samples. To guarantee consistency and capture minor AMD differences, the dataset is rigorously preprocessed. This large dataset serves as the foundation for training the RF and CNN models, which allows them to identify complex patterns unique to AMD. The models are put through extensive training to get a thorough understanding of the complexity of AMD, and rigorous validation procedures are used to refine parameters and improve flexibility. Diverse performance indicators highlight the advantages and disadvantages of the model in an examination of an unpublished dataset that simulates real-world situations. The CNN performs admirably, with a 97.08% diagnosis accuracy for AMD, demonstrating its superior ability to comprehend complex disease patterns. On the other hand, AMD phases and severity levels are effectively distinguished by the RF model. This study represents a significant advancement in the treatment of agricultural diseases by developing accurate, automated methods for quick AMD detection. Combining state-of-the-art DL with conventional models could strengthen crop protection, allow for prompt interventions, and maximize resource allocation for sustainable farming methods.