Uso de Ensemble Learning para controle de qualidade de hortaliças

The development of technology for agriculture has proven essential for increasing efficiency in plant disease detection and, consequently, improving agricultural production. In this context, this work presents a proposal focused on implementing a plant disease classification system using machine learning techniques. The main objective was to build a model capable of identifying specific diseases in potato, corn, and tomato crops, assisting farmers in managing their crops. Different classification models, such as Decision Tree, Random Forest, SVC, KNN, Gradient Boosting, and Stacking Learning, were implemented and tested to select the most accurate model for this application. The system was developed in Python, using the Scikit-Learn library, and the models were evaluated based on metrics like Mean Absolut Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and accuracy. The results indicated that the Stacking Learning model achieved the best performance in disease classification, standing out for its superior accuracy. Additionally, a React Native interface was created to facilitate the system’s use on mobile devices. Upon the conclusion of this work, an automated plant disease diagnosis tool was proposed, offering valuable support for agriculture and more efficient crop management.