Polarimetric imaging method for surface quality evaluation of a liquid metal pool obtained during welding
Affiliation auteurs | !!!! Error affiliation !!!! |
Titre | Polarimetric imaging method for surface quality evaluation of a liquid metal pool obtained during welding |
Type de publication | Journal Article |
Year of Publication | 2020 |
Auteurs | Kechiche AZanzouri, Aubreton O, Mathieu A, Mannucci A, Stolz C |
Journal | OPTICAL ENGINEERING |
Volume | 59 |
Pagination | 100501 |
Date Published | OCT |
Type of Article | Article |
ISSN | 0091-3286 |
Mots-clés | Computer vision, liquid metal, optic, polarimetric imaging, stokes parameters, welding process |
Résumé | Conventional methods used to inspect the molten pool surface induced by arc welding processes are based on post-welding evaluation using destructive or semi-destructive testing methods. Many attempts were performed to develop alternative and nondestructive tests. In this context, a polarimetric vision methodology is evaluated. The polarimetric parameters are measured from the thermal radiations emitted by the hot liquid metal at a wavelength within an arc plasma blind spectral window. The Stokes parameters and their segmentation within a Gaussian mixture model allow to discriminate artifacts at the surface of the liquid metal. After weld pool solidification, the use of scanning electron microscopy and energy dispersive spectroscopy allow to identify the artifacts, i.e., floating alumina particles. (C) 2020 Society of Photo-Optical Instrumentation Engineers (SPIE) |
DOI | 10.1117/1.OE.59.10.100501 |