Implementation of vision system for defects classification method using international standard welding imperfection
Defects are ones of the problems that should be minimized by any manufacturing company. Defects occurrence will decrease the performance of a production system. Defects typically occur in many kinds of product especially in manufacturing field. This research proposes an automated inspection of weldi...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/23333/1/Implementation%20Of%20Vision%20System%20For%20Defects%20Classification%20Method%20Using%20International%20Standard%20Welding%20Imperfection.pdf http://eprints.utem.edu.my/id/eprint/23333/2/Implementation%20of%20vision%20system%20for%20defects%20classification%20method%20using%20international%20standard%20welding%20imperfection.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Defects are ones of the problems that should be minimized by any manufacturing company. Defects occurrence will decrease the performance of a production system. Defects typically occur in many kinds of product especially in manufacturing field. This research proposes an automated inspection of welding defects particularly the surface defects that are normally inspected manually. In order to inspect the welding product, Metal Inert Gas (MIG) welding has been chosen to be tested using three types of gases which are Carbon Dioxide (CO2), Argon (Ar) and gas mixture (combination of CO2 and Ar). The selection of the three types of gases is because the three types are the most widely used gases in the industry. Currently,many of the companies in Malaysia are implementing manual visual inspection conducted by operators. As welding is a highly repetitive process, the operators will have high hazard exposure.In order to improve the working condition, vision system is proposed. This research proposes a classification of the defects method that involves with image enhancement, image filtering and image segmentation. The classification method is involve with decision tree. A proposed of decision tree is lead to the image determination. In image filtering method, an improved method is applied to get a better picture of the image. The peak signal to noise ratio (PSNR) and mean square error (MSE) is calculated. The result obtained shows that the system is capable to detect welding defects successfully. According to the result, percentage error rate for both situation of the noise shows that the result is less than 30%.In addition, it shows that the most of the result is same with the expertise. Thus, the system is success for the classification of welding defect. |
|---|