Weld seam detection using computer vision for robotic arc welding

Visual identification of narrow welding seams is difficult to achieve in robotic systems especially for ferrous materials. This paper describes an algorithm for detecting the weld seam in a butt-joint configuration for robotic Arc Welding of mild steel materials using computer vision. The proposed method can automatically subtract the background from the images obtained from a robot mounted camera system. Reliable seam identification can then be achieved for welding of ferrous materials. Experimental results have shown that this method is capable of detecting both straight and curved welding seams without prior knowledge of the location of the seam. It is shown that this method is robust to be used in an industrial setting.