Vinepredict - AI-Driven Powdery and Downy Mildew Forecasting Models

The objective of our project is to train artificial neural network models, namely sequence models, to predict downy and powdery mildew using pre-existing (5 years) and newly collected training data presenting site-specific differences in disease occurrence. Our research approach considers not only weather parameters but also vine development and vegetation cover. The susceptibility to both diseases changes during the development of the vines. The output of the models can provide valuable information for winegrowers on when to apply fungicides, thereby reducing their use.