This study investigates the application of artificial intelligence, specifically deep learning-based image processing techniques, for the detection of grapevine diseases in agricultural settings. Leveraging a publicly available dataset from Kaggle, the project focuses on classifying grape leaves as either healthy or affected by one of three common diseases: Black Rot, Esca (Black Measles), and Leaf Blight. Three machine learning models were developed and evaluated: Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), and Transfer Learning. Each model was trained and tested using the same dataset to ensure a fair comparison. Among the models, the CNN achieved an accuracy of 97.40%, while the DNN model showed significantly lower performance at 31.41%. Transfer Learning outperformed the others, reaching a peak accuracy of 98.84%. The results underscore the potential of deep learning, particularly transfer learning, in automating disease identification processes in viticulture. Such AI-driven systems can enhance precision agriculture by enabling early detection and prompt intervention, ultimately contributing to improved crop yield and quality.
Hasani et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: