Publicação

Machine Learning for precision viticulture: Focusing on Plant Health

Ver documento

Detalhes bibliográficos
Resumo:Precision Agriculture (PA) represents a significant advancement in agricultural practices, focusing on efficiency and sustainability through the integration of cutting-edge tech- nologies. This dissertation introduces a comprehensive software solution, "SmartData", which is integrated into the existing "AgriDash" application to enhance the management and analysis capabilities in agriculture. SmartData leverages satellite imagery to generate intuitive maps showcasing vegetation, water, and humidity indices, crucial for effective crop management. Additionally, it incorporates a machine learning model for the accu- rate classification of plant diseases from images, facilitating timely and effective disease management. The system also includes a geolocated disease marker system, providing precise location and detailed information about plant health issues, and integrates meteorological data from local weather stations to aid in decision-making based on weather conditions. These functionalities not only enhance AgriDash but also align with the goals of PA by improving productivity and sustainability. The dissertation outlines the development of a disease classification system, the cre- ation of advanced visualization tools for interpreting satellite data, and the integration of real-time meteorological data to enhance agricultural decision-making. These innovations have been integrated into the AgriDash application, improving its utility for sustainable farming practices. The relevance and effectiveness of these solutions are further under- scored by the acceptance of a related paper for presentation at the 15th International Conference on Ambient Systems, Networks and Technologies (ANT’24).
Autores principais:Madeira, Miguel Ângelo Lage
Assunto:Precision Agriculture SmartData AgriDash Machine Learning Satellite Imagery Disease Classification
Ano:2024
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade Nova de Lisboa
Idioma:inglês
Origem:Repositório Institucional da UNL

Registos relacionados

Registos relacionados