Development of a Machine Learning model to detect diseases in vineyard

The activity of the thesis will focus on the development of a Decision Support System for diseases detection in vineyard based on proximal and remotely sensed spectral data, in the context of Precision Agriculture (PA) framework. In particular, spectral signatures of vine leaves will be acquired with a portable spectroradiometer at several timing during the growing season, and specific data processing will be developed and implemented in order to investigate the effect of flavescence doreé on the reflectance of leaves. Machine learning based method will be then trained in order to define a robust disease classifier. Finally, the possibility to extend the developed approach to remotely sensed by UAV will be investigated by feasibility study.