A visual sensing system for welding control and seam tracking in aluminum alloy gas tungsten arc welding

A welding control system was developed for automatic pulsed gas tungsten arc welding (GTAW) of aluminum alloy with root pass. The system consisted of a vision sensor that consisted of a charged-coupled device (CCD) camera and lenses, image processing algorithms, and a computer controller. Specially designed three-direction vision sensor was used to view the weld pool from top-back, top-front and back side. During welding, an image of the weld pool and its vicinity was captured when basic current of welding power. The top-front side part image was then processed to get gap size and welding direction for adjusting welding condition and seam tracking. The image process technique included degradation recover, automatic threshold segmentation, area threshold removing noise, thinning and Hough transform.