An active visual monitoring method for GMAW weld surface defects based on random forest model

• Published in: Materials research express Vol. 9; no. 3; pp. 36503 - 36514
• Main Authors: Zhu, Caixia, Yuan, Haitao, Ma, Guohong
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.03.2022
• Subjects: Accuracy; active vision; Algorithms; Automatic welding; Decision trees; Defects; Design defects; Feature extraction; feature point extraction; Gabor filters; Galvanized steel; Galvanized steels; Gas metal arc welding; image processing; Industrial robots; Laser beam welding; Machine learning; Monitoring; random forest; Real time; Spatial distribution; Surface defects; Surface properties; Weld defects

In the automatic manufacturing of robotic welding, real-time monitoring of weld quality is a difficult problem. Meanwhile, due to volatilization of zinc vapor in galvanized steel and complexity of welding process, the existence of welding defects greatly affects industrial production process. And re...