Automated visual inspection of friction stir welds using image segmentation algorithms
- Vieltorf F. ,
- Eck L. ,
- Bidlingmaier F. ,
- et al
- Vieltorf F. ,
- Eck L. ,
- Bidlingmaier F. ,
- Sigl M.E. ,
- Zens A. and
- Zaeh M.F.
Links
Repository
Description
Friction stir welding (FSW) is a solid-state welding process, particularly suitable for aluminum alloys. The process allows the production of welds with a high seam strength and low distortion. Despite these benefits, irregularities can occur during the process, leading to visible defects on the weld seam surface, such as toe flash and surface galling. Human visual inspection is often used to detect these defects after welding, which poses challenges because of its subjectivity and because of the low reproducibility. In addition, especially for large weld seams like in the aerospace sector, the surface characteristics can only be measured in a highly discretized manner. To solve these problems, an automated visual inspection system, which is presented in this article, was developed. The system uses image segmentation algorithms that enable the automatic acquisition, processing, and evaluation of seam surface features.
13th International Symposium, 21-23 May 2024, Session 7: Fundamentals, Paper 03
Related Records
