Tipo di tesi
Tesi di laurea magistrale
Titolo
Development of Machine Learning Models for Plant Disease Risk Assessment Using Hyperspectral Data
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
ARTIFICIAL INTELLIGENCE AND DATA ENGINEERING
Parole chiave
- Artificial Intelligence
- Decision Support Systems (DSS)
- hyperspectral sensing
- Integrated Pest Management (IPM)
- Machine Learning
- plant disease risk assessment
- Plasmopara viticola
- Precision Agriculture
Data inizio appello
26/05/2026
Consultabilità
Non consultabile
Data di rilascio
26/05/2096
Riassunto (Inglese)
The sustainability of modern viticulture is compromised by Plasmopara viticola and the ecological toll of prophylactic fungicides. Conventional Decision Support Systems rely on mechanistic weather models, but their hypersensitivity generates excessive false alarms, perpetuating chemical overuse. This thesis presents the wAIne project, a pioneering open-source Artificial Intelligence architecture that redefines disease forecasting through multi-modal data fusion. By synergizing hyperspectral imaging with microclimatic time-series, the system detects the invisible biophysical stress of plant tissue before macroscopic symptoms appear. A robust data engineering pipeline was engineered to harmonize heterogeneous sensors, filter chemical persistence, and conquer the severe class imbalances native to agricultural data. Utilizing a gradient boosting multi-output regressor, the architecture generates dynamic, multi-horizon infection probabilities.
This research delivers a scalable, interpretable diagnostic engine that bridges theoretical remote sensing and applied Precision Agriculture, propelling the sector toward Agriculture 5.0.
