Development of a workpiece classification system using an electro-pneumatic robot and machine vision
Yachel Rogério Mileski1; Camila Emer Sartori1; MAria Cristina Minusso Kopacek1; Fabiano Disconzi Wildner1; Heraldo José de Amorim1
1 UFRGS
doi:10.20906/CPS/COB-2015-2264
Resumo
With the growing movement towards automation in industrial manufacturing and distribution, machine vision systems are quickly being adopted in different fields. Among other areas, these systems are applied in astronomy, medicine, logistics and manufacturing. Vision systems are used in different configurations, however, most applications use an optical system for image acquisition, a set of computer scripts for image processing and data interpretation and a hardware for computing data. In this work, the goal lies in the development of a methodology for part recognition that allows both shape and size identification, aiming the organization of production through a 3-axis electro-pneumatic robotic manipulator. For recognition of geometric form, both centroid and the moment of inertia of parts are used; size can be compared through direct pixel comparison. After determination of part size and form, a positioning algorithm is applied on the robot, thus allowing its positioning for workpiece classification.
Palavras-chave: Machine vision; image processing; image recognition; electro-pneumatic robot