Ciencias Agropecuarias

URI permanente para esta comunidadhttp://repositorio.uta.edu.ec/handle/123456789/870

Examinar

Filtros

20251

Ajustes

Autor/a: search.filters.author.Alomalisa Machado, Christian IvanTema: search.filters.subject.INTELIGENCIA ARTIFICIAL

Resultados de la búsqueda

Mostrando 1 - 1 de 1
  • Miniatura
    Item type:Ítem,
    Inteligencia artificial para la detección de dos enfermedades en manzana (Malus domestica, var. Emilia)
    (2025-08)
    Alomalisa Machado, Christian Ivan
    ;
    Muñoz Espinoza, Manolo Sebastián
    Early detection of diseases in agricultural crops is essential to ensure sustainable and profitable production. In this context, artificial intelligence (AI) has become a key tool for the development of automated plant health diagnosis systems. This thesis aims to develop an AI-based model for the detection of two common diseases in apple trees: powdery mildew (Podosphaera leucotricha) and apple scab (Venturia inaequalis), using digital images of leaves. To achieve this, a database of apple leaf images (cv. Emilia) was created, classified into three categories: healthy, with mildew, and with scab. The images were acquired, preprocessed, and annotated using the Roboflow platform. Subsequently, an object detection model based on YOLOv12 was trained to identify and visually locate affected areas in each image. The performance of the model was evaluated using metrics such as precision, sensitivity, and specificity, yielding promising results. The model showed optimal performance in identifying healthy leaves and those affected by scabs, along with a good balance between sensitivity and precision in detecting mildew. These findings demonstrate the feasibility of using deep learning models for the automated detection of foliar diseases in fruit trees. This study contributes to the development of accessible and scalable technological solutions for the efficient monitoring of diseases in apple crops, supporting informed decision-making in integrated pest and disease management.