Anzeige
Erschienen in:
2020 | OriginalPaper | Buchkapitel
Welding Deviation Extraction during K-TIG Welding Based on K-Means Clustering
verfasst von : Baori Zhang, Yonghua Shi
Erschienen in: Transactions on Intelligent Welding Manufacturing
Verlag: Springer Singapore
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
During K-TIG welding, due to the strong arc and the narrow seam, it is difficult to accurately identify the deviation of the welding, which would affect the K-TIG automatic welding. In this paper, a novel welding deviation visual inspection method is proposed. A high dynamic range camera is used to capture the welding images. Then, K-means-based image segmentation algorithm is used to automatically divide an image into arc region and seam area. An optimal path method is proposed to extract the feature points on the electrodes from the arc region image and get the position of the electrode tip points. The seam area is processed by a refinement algorithm to obtain the weld position, and finally, the deviation between the electrode tip and the seam is obtained. The experiment proves that the proposed algorithm can realize the automatic identification of K-TIG welding deviation.