Hitsaussauman paikan mittausmenetelmän kehittäminen laadunvarmistusautomaatioon
Becker, Kalle (2016)
2016
Automaatiotekniikan koulutusohjelma
Teknisten tieteiden tiedekunta - Faculty of Engineering Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2016-06-08
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201512281865
https://urn.fi/URN:NBN:fi:tty-201512281865
Tiivistelmä
The card industry uses machine vision in increasing amounts to perform quality inspections on their products. The commercial software used in the making of this thesis is used in the in-line inspection process of different car manufacturers. The software is used to analyze and document data obtained by a machine vision system. This thesis focused on developing the inspection of welding seams by automating the work flow used in defining welding seams from images. The definition had previously created an unwanted amount of work for the end user and slowed down the usability of the software. By improving the functionality of the software it was possible to partly improve the process for quality assurance and to minimize human error created in the defining of welding seams.
In this thesis the goal was to create an automatic segmentation tool for weld seams and at the same time lower the amount of manual adjustment required to run the software. Through the implementation of the new functionality we were able to research the sufficiency of the transformation method used on the weld seams while considering the requirements of quality assurance. The new functionality was tested by using documented measurement data to avoid the need to use any physical measurement system.
Results include examples obtained using the created tool. The used transformation for welding seams was insufficient considering the quality of point data that is available for the current software. To automatically define weld seams in the software the quality of data has to be improved or the transformation needs to be performed locally.
In this thesis the goal was to create an automatic segmentation tool for weld seams and at the same time lower the amount of manual adjustment required to run the software. Through the implementation of the new functionality we were able to research the sufficiency of the transformation method used on the weld seams while considering the requirements of quality assurance. The new functionality was tested by using documented measurement data to avoid the need to use any physical measurement system.
Results include examples obtained using the created tool. The used transformation for welding seams was insufficient considering the quality of point data that is available for the current software. To automatically define weld seams in the software the quality of data has to be improved or the transformation needs to be performed locally.