Estimación del nivel de impurezas en muestras de aceite
Archivos
Fecha
2021-09-06
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Jaén: Universidad de Jaén
Resumen
La producción de aceite de oliva virgen (AOV) es una importante actividad económica que debe mantener su
competitividad en un entorno de economía global. Uno de los aspectos que determinan la calidad en la extracción del
aceite, es la presencia de impurezas insolubles. El objetivo principal de este trabajo fue el desarrollo e implementación
de un sistema de visión por computador para determinar el nivel de impurezas presentes en muestras de aceite de
oliva virgen en condiciones de laboratorio. Para ello se estudiaron cuatro vectores de entrada diferentes, derivados
del histograma de los canales de los espacios de color RGB, HSV y CIELAB. Además, se aplicó un método de extracción
de características antes de la clasificación. El mejor resultado de clasificación se logró utilizando un método de
extracción de características Kernel Principal Component Analysis (KPCA) junto a un clasificador Random Forest, con
una precisión del 65,45%.
The production of Virgin Olive Oil (VOO) is an important economic activity that must remain competitive in a global economic environment. One of the aspects that determine the quality in oil extraction process is the content of insoluble impurities. The main objective of this work was the development and implementation of a computer vision system classify the content of impurities in virgin olive oil samples under laboratory conditions. For this purpose, three different input vectors constructed from the histograms of the channels of the RGB, HSV and CIELAB colour spaces were studied. Besides, a feature extraction was performed before classification. The best classification result was achieved using a Kernel Principal Component Analysis (KPCA) jointly with a Random Forest classifier, with an accuracy of 65.45%.
The production of Virgin Olive Oil (VOO) is an important economic activity that must remain competitive in a global economic environment. One of the aspects that determine the quality in oil extraction process is the content of insoluble impurities. The main objective of this work was the development and implementation of a computer vision system classify the content of impurities in virgin olive oil samples under laboratory conditions. For this purpose, three different input vectors constructed from the histograms of the channels of the RGB, HSV and CIELAB colour spaces were studied. Besides, a feature extraction was performed before classification. The best classification result was achieved using a Kernel Principal Component Analysis (KPCA) jointly with a Random Forest classifier, with an accuracy of 65.45%.