Diseño e implementación de un prototipo que permita la identificación por tipo y tamaño de la naranjilla y del tomate de árbol mediante visión artificial.

Abstract

The present research was to design and implement a prototype that allows the identification by type and size of naranjilla and the tree tomato, to achieve this, an artificial computer vision system based on neural networks was used. The prototype has three systems: the image acquisition system made up of a webcam and lighting for data capture; the artificial vision system uses the image processing method to carry out the identification by size and neural networks for identification by type, finally, the mechanical system is made up of a storage system and a conveyor belt. For the training of the Artificial Neural Network, the convolutional algorithm was used that through the acquisition of images trains the identifier, three categories were established: tree tomato, naranjilla, and unknown fruit, each one has a database. The algorithms programming was developed in Raspbian, with the Python programming language using OpenCV and TensorFlow libraries. The graphical interface was developed using the Tkinter library, allowing the user to control six buttons: start camera, take the photo, upload photo, type of fruit, size of fruit, and exit. Based on the tests carried out, it is established that the system works optimally with 95% when detecting the type of fruit. 93% to identify the size, the response time ranges between 0.3 and 1.3 seconds. It is concluded that the prototype achieves the identification of the fruit by their type and size with a minimum margin of error. It is recommended to reduce the response time for identification using a graphics card.