Preverjanje kakovosti izdelka s strojnim vidom v procesu montaže

Primožič, Rok (author), Šimic, Marko (mentor), Herakovič, Niko (comentor)

Naloga prikazuje razvoj sistema s strojnim vidom za prepoznavanje pravilnosti vzorca na paleti v montažnem proizvodnem procesu. Glavni namen je preveriti možnost uporabe nizkocenovnih komponent, ki bi še zagotavljale dovolj kakovosten sistem strojnega vida za vpeljavo na proizvodno linijo. Za sistem strojnega vida je bila zato izbrana nizkocenovna kamera, s katero smo najprej izvedli eksperimentalno analizo za prepoznavanje vzorcev pri uporabi naravne svetlobe. Določili smo osnovne parametre, kot so: oddaljenost kamere od objekta, parametre kamere in osnovne dimenzije sistema strojnega vida. Za izvedbo preizkusov smo napisali programsko kodo za prepoznavanje vzorcev. Testi so pokazali, da za kakovosten proces zaznavanja vzorcev potrebujemo zaprto komoro z ustrezno osvetlitvijo, da izločimo vpliv okoliške naravne svetlobe, ki povzroča odboje svetlobe od opazovanih objektov. Nadalje smo izvedli teste za oceno kakovosti novega sistema strojnega vida na več različnih testnih vzorcih in določili optimalne parametre kamere in osvetlitve za doseganje 100 % zaznavanja vzorcev.

• Language: Slovenian
• Keywords: montaža, strojni vid, prepoznavanje, kontrola, kamera, osvetlitev, eksperimentalna analiza
• Work type: Bachelor thesis/paper
• Organization: FS - Faculty of Mechanical Engineering
• Year: 2018
• PID: 20.500.12556/RUL-101273
• Publication date in RUL: 19.05.2018

Secondary language

• Language: English
• Title: Testing the quality of products with machine vision in assembly process

Abstract

My task was to show the progress of system with machine vision for recognition of the regularity of the sample carried out on pallet in the production process. The main task is to check the possibillity of usage of low-cost components, which can substitute the professional components. For the purpose of the tests, a low-cost camera was selected, which represents the basic component for the execution of the control chain. Analysis of the quality checks was implemented in a special chamber, in which we subsequently installed a low-cost camera and elements of lighting. On the basis of the selected configuration of the elements of the control chain the software code for processing the test patterns has been purposefully designed. In the final stage we tested the adequacy of the functioning of the whole system based on the software code. We have examined the adequacy of the operation of the control system, on several different test samples and therefore confirmed the proper functioning of the system for checking the quality of the products.

• Keywords: assembly, machine vision, recognition, testing, camera, lightning, experimental analysis