Sistema de procesamiento de imágenes y su influencia en la gestión del consumo energético residencial

Currently, image processing is a technology that is in constant use in both industrial and residential sectors. This technology consists of analyzing the characteristics of the images and obtaining information that can be processed by a computer to perform tasks such as detecting people or objects a...

Descripció completa

Guardat en:
Dades bibliogràfiques
Autor principal: Guillén Sánchez, José Gregorio (author)
Altres autors: Del Valle Villacís, Bryan Alejandro (author)
Format: bachelorThesis
Idioma:spa
Publicat: 2019
Matèries:
Accés en línia:http://repositorio.uteq.edu.ec/handle/43000/3928
Etiquetes: Afegir etiqueta
Sense etiquetes, Sigues el primer a etiquetar aquest registre!
Descripció
Sumari:Currently, image processing is a technology that is in constant use in both industrial and residential sectors. This technology consists of analyzing the characteristics of the images and obtaining information that can be processed by a computer to perform tasks such as detecting people or objects and following their movement. Therefore, this research project focuses on performing an image processing system applied on a Raspberry Pi 3B + that influences the management of residential energy consumption. The constant project of an electrical control circuit that is responsible for detecting people and managing the consumption of electrical energy from the modification of light intensity and temperature regulation. The constant three-stage project, in the first stage the characteristics of the optimal elements for the system hardware will be determined, in the second stage the algorithm that allows the detection of people will be developed, while in the third stage it was carried out The implementation of the complete system in residences with regulated environments. Finally, control system tests are carried out in controlled environments and the results are quantified using the confusion matrix. The results obtained from the matrix metrics that the system has a detection rate of 83% and an error rate of 17%, which are acceptable percentages even though the processing is in real time. Also, the difference in energy consumption between the conventional and the automated system.