An Image Processing Based Method for Welding Defect Detection under Complex Environment

J. J. Zhu, F. Wang

School of Electrical & Automatic Engineering, Changshu Institute of Technology, Changshu, 215500 China

Cite this paper
J. J. Zhu, F. Wang, “An Image Processing Based Method for Welding Defect Detection under Complex Environment”, Journal of Mechanical Engineering Research and Developments, vol. 39, no. 2, pp. 513-520, 2016. DOI: 10.7508/jmerd.2016.02.030

ABSTRACT: Recognition and identification of weld defects dimensional position by computer vision is a key technology for practical industrial applications, especially under complex background. Aiming at this requirement, a method for weld defect detection under complex environment is proposed. Firstly, pre-location process is executed, in which filter technology, morphological technology, some geometrical features and edge detection are used to obtain the object in the vision. Secondly, line scan and a cluster algorithm are applied to find out the defect candidates. In the end several features are adopted to find out the real weld defect. The experimental results show that our method is effective in detecting weld defect under complex environment.

Keywords : Welding detection; Image process; Morphology, Cluster.

[1] X.G. Zhang, J. J. Lin. “Research of Image Processing and Defect Recognition for Industrial Radiographic Weld Inspection.” Journal of East China University of Science & Technology 30.2(2004): 199-202.
[2] R.Vilar, J.Zapata,  R. Ruiz, (2009). An automatic system of classification of weld defects in radiographic images. NDT and E International, 42(5), 467–476.
[3] Valavanis, Ioannis,  D. Kosmopoulos. “Multiclass defect detection and classification in weld radiographic images using geometric and texture features.” Expert Systems with Applications 37.12(2010): 7606-7614.
[4]  V. Lashkia, “Defect detection in X-ray images using fuzzy reasoning.”Image & Vision Computing 19.5(2001): 261-269.
[5] Q.M. Shen, J. Gao, L. I. Cheng. “Recognition of Weld Defect Types.” Journal of Xian Jiaotong University (2010).
[6] Nacereddine, N., Zelmat, M., Belaïfa, S. S., & Tridi, M. “Weld defect detection in industrial radiography based digital image processing.” In: Proc. of World Academy of Science Engineering & Technology (2005): 112.