MACHINE VISION IMPLEMENTATION FOR OFF-LINE MEASUREMENT OF WELD BEAD GEOMETRY IN API X65 PIPELINE STEEL

摘要

In this paper, machine vision is used for measurement of weld bead geometry in submerged arc welding of API X65 steel plates. First, different welding parameters including voltage, current and speed are set to perform the SAW with different heat inputs. Then, different etchants are used to find an appropriate solution for the best revelation of weld metal in the cross section of welded plates. Finally, weld bead parameters (including weld width, depth, height, area and boundary) are measured using machine vision technology and image processing techniques. The comparison of machine vision results with those from manual measurements showed the higher accuracy of the proposed vision system and its faster performance in accurate measurement of weld bead geometry.