Sistema de control de calidad utilizando redes neuronales para la clasificación del estado de la granadilla

• Autores: Cuya Marzal, Bryan Cristhian Martín, Ramos Lugo, Martín Guillermo
• Formato: tesis de grado
• Fecha de Publicación: 2020
• Institución: Universidad de Lima
• Repositorio: ULIMA-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.ulima.edu.pe:20.500.12724/13854
• Enlace del recurso: https://hdl.handle.net/20.500.12724/13854
• Nivel de acceso: acceso abierto
• Materia: Control de calidad; Plagas del campo; Redes neuronales (Informática); Passion fruits; Quality control; Agricultural pests; Neural networks (Computer science); Granadilla; https://purl.org/pe-repo/ocde/ford#2.02.04

This paper is about the most critical issue that we are living right now, the propagation of diseases with high probability of spread, and a potential solution to improve the current quality process after the collection of fruits from business related with agriculture that had been impacted by several actions from Governments and their ministries. Also, it develops factors of desirability, feasibility and expected benefits with the implementation of this project to the final client who are the independent farmers through a low-cost alternative oriented to the requirements, like the accomplishment of the protocols and the optimization of principal activities during quality assurance. This alternative uses a disruptive technology as internet of things for connection and communication between a computer and a camera that will take photos and will send wirelessly with the purpose of being evaluated in a neural convolutional network model that can classify the production of a kind of Peruvian fruit, granadilla, as valid or invalid while the protocols are being fulfilled and making business improves. Finally, it presents considerations and lessons learned along the development of four tests of solution in the generation of the network and the shipment of information between devices.