Detection of Weld Cracks Using Magneto-optical Imaging

High strength steel weld cracks are recognized by magneto-optical imaging. The background and basic principle of micro weld crack detection using magneto-optical imaging(MOI) technology are discussed. Based on fuzzy set theory, continuous fuzzy enhancement algorithm is used to improve the differentiation degree on weld cracks and melting zone, and solve unevenness of crack magneto-optical imaging of high strength steel surface. An identification model of laser welding cracks is established by using principal component analysis(PCA) method and support vector machine (SVM). Experimental results show that the proposed method could effectively extract the characteristics of weld crack and improve the accuracy of weld crack detection.