A vision system for automated crack detection in welds

The authors describe the development of a vision-based system for automated inspection of surface cracks in materials. The magnetic particle inspection method is used to reveal cracks in ferromagnetic materials. The vision system consists of an IBM AT compatible computer, a framestore and a camera. Details of the preparation procedure and the software algorithms for crack recognition are discussed. Results are presented for transverse and longitudinal cracks in welds used to fabricate structures.