The increasing use of automated laser welding processes causes high demands on process monitoring. This work demonstrates methods that use a camera mounted on the focussing optics to perform pre-, in-, and post-process monitoring of welding processes. The implementation uses machine learning methods. All algorithms consider the integration into industrial processes. These challenges include a small database, limited industrial manufacturing inference hardware, and user acceptance.
Julia Hartung
Künstliche Intelligenz Maschinelles Lernen Laserschweißen Hairpin Technologie semantische Segmentierung Qualitätssicherung machine learning quality assurance laser welding hairpin technology semantic segmentation stacked dilated U-Net CNN