A robust welding seam identification method

As an automatic welding process may experience some disturbances caused by, e.g., splashes and/or welding fumes, misalignments/poor positioning, thermally induced deformations, strong arc lights, diversified welding joints/grooves, etc., precisely identifying the welding seam has an great influence on the welding quality achieved. In this paper, a robust method for identifying this seam is proposed. Firstly, after a welding image obtained from a/the structured-light vision sensor is filtered, in a sufficiently small area, the extended Kalman filter (EKF) is used to search for the/its laser stripe in order to prevent possible disturbances. Secondly, to realize the extraction of the profile of welding seam, the least square method is used to fit a sequence of centroids determined by the scanning result of columns displayed on the tracking window. Thirdly, this profile is then qualitatively described and matched using a proposed character string method. Finally, the advantages of this method are compared with those of other approaches through repeated experiments.