Spremljanje procesa montaže z uporabo strojnega vida

magistrsko delo

Peter Lampret (Author), Simon Klančnik (Mentor), Tadej Peršak (Co-mentor)

Abstract

Magistrsko delo obravnava uporabo strojnega vida v procesu montaže. V prvem delu naloge imamo obravnavane teoretične osnove strojnega vida s poudarkom na sledenju objektov. V nadaljevanju naloge je prikazano oblikovanje montažnega mesta, ki omogoča spremljanje procesa montaže s strojnim vidom. Nato pa je prikazan razvoj programa za sledenje komponent in preverjanje pravilnosti montaže na tlačilki za žoge. Program je izdelan v programskem okolju MATLAB, najprej je prikazano, kako iz videa dobimo temeljne podatke, ki jih uporabimo za učenje detektorjev, ti morajo prepoznati vse komponente tlačilke in pravilnost sestave. Na koncu sledi prikaz delovanja programa in rezultati. Izkaže se, da program dobro prepozna pravilnost sestave, nekoliko več težav pa ima s sledenjem komponent čez celoten posnetek.

Keywords

strojni vid;sledenje objektov;montaža;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2022
Typology:	2.09 - Master's Thesis
Organization:	UM FS - Faculty of Mechanical Engineering
Publisher:	[P. Lampret]
UDC:	004.93:658.515(043.2)
COBISS:	151529987
Views:	224
Downloads:	29
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Monitoring the assembly process using machine vision
Secondary abstract:	The master’s thesis discuses the use of machine vision in the assembly process. The first part deals with the theoretical fundamentals of machine vision with emphasis on object tracking. Furthermore, the thesis illustrates the design of assembly workplace which enables monitoring of the assembly process with machine vision. This is followed by the development of the program for object tracking and checking the correctness of the ball pump assembly. The program was made in MATLAB software. First of all, the process of how we get ground truth data from a video, for object detector training, is shown. The detectors must then recognize components and correctness of assembly. Lastly, the program operation and results are shown, along with the revelation, that the program recognizes the correctness of assembly well, but it has slightly more problems with component tracking over the entire video footage.
Secondary keywords:	machine vision;object tracking;assembly process;MATLAB;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za strojništvo, Proizvodne tehnologije in sistemi
Pages:	1 spletni vir (1 datoteka PDF (IX, 50 f.))
ID:	16364780