Clasificador de productos agrícolas para control de calidad basado en Machine Learning e Industria 4.0.

Abstract

Currently, empirical techniques in Ecuadorian agricultural production are not enough to achieve quality changes with food safety standards and thus meet the demand of an international market. This work presents a system capable of supervising, classifying, and controlling the quality of products in the agricultural sector, by applying soft computing techniques and machine learning algorithms that affect the identification of images in real time. The research will implement classification algorithms of nearest K neighbors to label the products according to their quality and send the reports in real time to a web application using the MQTT protocol. The employed products in this study were bananas, oranges, green bananas, and apples. The obtained results allow to determine the minimum number of images required for training the identification models and the identification error rates during the validation stage.