Sistema de monitoreo de temperatura y humedad con determinación del estado de cosecha para cultivo de forraje verde de Maíz Hidropónico

This project covers the topic of pattern recognition to know the stage of development of hydroponic forage and thus determine the optimal harvesting time. Specifically, it develops two variants: initially, the use of Image Processing approaches related with Machine Learning, and after, the use of deep learning in the disclosure and identification of patterns in the production process of hydroponic green fodder; the project proposes to implement a greenhouse prototype for the production of hydroponic green fodder, by controlling the variables (light intensity, water, temperature, humidity); the control of these variables is related to precision agriculture that has been utilized for collecting and processing crop data. Process control methods based on embedded systems, image recognition and learning through neural networks has been utilized with the goal of obtaining ideal values of these variables to carry out the production process of hydroponic green fodder and at the same time, thanks to self-learning and monitoring of variables, each time the process is carried out, a better result will be obtained.

• Autores: Cerquera Carvajal, Dilan Amaury
• Tipo de recurso: Tesis
• Fecha de publicación: 2023
• Institución: Universidad Antonio Nariño
• Repositorio: Repositorio UAN
• Idioma: spa
• OAI Identifier: oai:repositorio.uan.edu.co:123456789/9812
• Acceso en línea: https://repositorio.uan.edu.co/handle/123456789/9812
• Palabra clave: Agricultura de precisión; Aprendizaje automático; Automatización; Hidroponía; Invernadero; Reconocimiento de Patrones; Precision agriculture; Automatic learning; Automation; Hydroponics; Greenhouse; Pattern recognition.
• Rights: openAccess
• License: Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)