Sistema de clasificación del cacao en la prueba de corte implementando procesamiento digital de imágenes

Abstract

Cocoa cultivation in Colombia is fundamental in the country's agricultural development, with thousands of families depending on this activity. The process of obtaining and selecting the cocoa bean consists of several crucial steps to ensure the quality of the final product. In particular, the cut test, which classifies the beans according to their color and degree of fermentation, is a key stage. This test is carried out manually by experts, which limits the objectivity and consistency of the results, affecting the producers and the quality of the cocoa. The dependence on the experience and skill of an expert in the classification of cocoa in the cutting test generates inconsistencies and difficulties for producers. Furthermore, the lack of accurate feedback up to the marketing stage especially affects small producers. The need then arises to find a solution that automates the determination of cocoa quality in this test, improving the consistency of the results and providing early and accurate feedback. This will require the design and construction of a prototype system based on artificial vision, which captures images of cut cocoa beans and processes them using parameters established by the NTC 1252:2021 standard. The proposed solution consists of developing a prototype system that uses artificial vision to automatically determine the quality of cocoa in the cutting test. This system will capture images of the cut grains and process them using an algorithm based on parameters established by the NTC 1252:2021 standard. The aim is to standardize quality control processes, improve the consistency of results and provide a comparison with the opinion of human experts. The design will include a structure that allows stable image capture, taking into account lighting changes. The estimated cost of the project is around $400 with an execution time of six months. The implementation of this solution has the potential to improve efficiency and quality in the classification of cocoa in the cutting test, benefiting both producers and the industry in general.