Use of Machine Learning Implementing Convolutional Neural Networks to Classify Export Markets for Persian Lime

Abstract

Mexico is the world's second largest producer of Persian lime; however, most producers of this citrus fruit cannot export on their own because they do not know the standards required to do so. As a result of this situation, they have to resort to intermediaries to carry out the fruit selection process, which causes them to lose a certain percentage of their profits. Therefore, the objective of this research is to generate a Machine Learning model with the implementation of convolutional neural networks to identify the different characteristics of the Persian lime and thus know to which market it should be exported. In this sense, the results obtained show that the model achieved an accuracy of 97% in its classification of the types of export markets established.