Author: ANTONIO VAROVIC
Company/Institution: University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture (FAMENA)
Country: Croatia
e-mail: varovicantonio@gmail.com; av230244@stud.fsb.hr
web: www.fsb.unizg.hr
Mentor: Prof. dr. sc. ZELJKO SITUM
e-mail: zeljko.situm@fsb.unizg.hr
An automated machine for sorting bad beans has been developed, using advanced neural network and machine vision technologies. This innovative system enables efficient sorting of beans, eliminating bad beans without the need for manual intervention. The system only needs to deliver the beans to the container, and begins the automated process of dosing and sorting beans according to quality. A comparison with traditional industrial busbars reveals significant advantages of this cheap machine. While industrial buses are much more expensive and unavailable for wider use, the developed automated machine provides an economical and efficient solution for bean sorting with minimal effort from the machine operator. The key parts of this automated machine include: a Controllino microcontroller that enables precise control of the system, a pneumatic Matrix valve that controls the process of ejecting bad grains, and a camera that allows the machine vision program to identify grain irregularities. Using sophisticated neural network algorithms, this machine precisely sorts bad beans and removes them from the conveyor belt with the help of compressed air at the output of a pneumatic valve. This system combines advanced technologies with practical application in agriculture.