Implementación de un clasificador de enfermedades en las plantas a través de técnicas de aprendizaje profundo

Abstract

Abstract: The primary objective of this master's thesis is to facilitate the detection and prediction of diseases that can affect plants, initially focusing on those that impact coffee crops. To achieve this goal, tests were conducted using the deep learning frameworks Yolov8 and YoloV11, developed by Ultralytics. The interface model was developed virtually using Python for the Windows operating system. For the interface to work, several tools were used, including OpenCV, which helped process images, allowing labels to be added and an advanced confidence level to be displayed. The Yolo programming language allowed us to use a pre-trained model, while Tkinter was used to design a simple interface with buttons and options for selecting images. Finally, the Os library helped us manage the necessary paths, resulting in software capable of predicting diseases that coffee plant leaves can suffer.