Simulation of an Automated Sorting System for Peruvian mangoes based on computer vision

• Autores: Rivera Rojas, Michael Cristhian, Yanarico Estrella, Maydelson Valentin, Ortiz Zacarias, Jhon Rodrigo, Quintanilla Mosquera, Iraiz Lucero, Valenzuela Lino, Yadhira Samhira, Coaquira Rojo, Carlos Alberto
• Formato: tesis de grado
• Fecha de Publicación: 2023
• Institución: Universidad Continental
• Repositorio: CONTINENTAL-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.continental.edu.pe:20.500.12394/14235
• Enlace del recurso: https://hdl.handle.net/20.500.12394/14235; https://doi.org/10.1109/ICOIACT55506.2022.9972217
• Nivel de acceso: acceso abierto
• Materia: Agricultura; Automatización; Exportación; https://purl.org/pe-repo/ocde/ford#2.03.01

Agriculture is one of the most important economic activities in Peru. Furthermore, the Peruvian mango is the most important fruit as an export product on the European market due to its good quality. Three varieties of mango are cultivated in Peru: Haden, Kent and Tommy Atkins, each of them presents different characteristics; however, the production of this fruit is affected because manual processes are still used for its production, which generates a delay in the process due to the inaccuracy of the labour. The present study aims to develop the simulation of an automated grading system for Peruvian mangoes based on computer vision. On the other hand, to establish the process development, the transfer learning principle was analysed and a flow diagram of the system was made. The simulation of the system environment was obtained through the Factory IO software, together with the image processing through the Matlab software, in which the characteristics of the three mango varieties were introduced; therefore, with this information, the system was able to carry out the mango selection. Finally, it was concluded that the presented design optimises the selection and storage time of the mangoes, as well as automating the labour in the process.