At present, the welding of pipes with large diameters is a manual process. Automation of the welding process is necessary for the sake of consistent weld quality and improvement in productivity.
In this study, two vision sensors, based on the optical triangulation, were used to obtain information for seam tracking and detecting weld defects. Through utilization of the vision sensors, noises were removed, images and 3D information obtained and the position of feature points detected. The aforementioned process provided seam and leg position data, calculated the magnitude of the gap, fillet area and leg length and judged weld defects by ISO 5817.
Noises in the images were removed by using the gradient values of the laser stripe's coordinates and various feature points were detected by using an algorithm based on the iterative polygon approximation method. This algorithm can be applied to bell end, butt and fillet joints. Since process time is very important, all the aforementioned processes should be conducted during welding. By using these methods we could finish all the processes fast and accurately.
Good weld quality was obtained through the use of a front vision sensor, which tracked the seam and a rear vision sensor, which detected weld defects. Further research could be conducted into the correction methods for weld defects.