Materials Science

, Volume 49, Issue 4, pp 469–477 | Cite as

Influence of Primary Processing on the Segmentation of X-ray Images of Welds

  • R. A. Vorobel
  • I. B. Ivasenko
  • T. S. Mandzii
  • V. V. Botsyan

We study the influence of preliminary processing of digital X-ray images by their median filtering and brightness normalization by nonsharp masking and contrast enhancement on the basis of a logarithmictype model on the results of testing. Examples of experimental detection of defects in welds of various types are presented.


Weld Filtering Contrast enhancement Segmentation of defects Logarithmic processing of images 


  1. 1.
    J. Shao, D. Du, H. Shi, et al., “Automatic weld recognition and extraction from real-time X-ray images using quadratic curve fitting and multi-order differences analysis of intensity profile,” Insight, 53, No. 10, 562–569 (2011).CrossRefGoogle Scholar
  2. 2.
    J. Hassan, A. Majid Awan, and A. Jalil, “Welding defect detection and classification using geometric features,” in: Proc. of the IEEE 10th Internat. Conf. on the Frontiers of Information Technology (December 17–19, 2012, Islamabad), Islamabad (2012), pp. 139–144.Google Scholar
  3. 3.
    M. Kroetz, T. M. Centeno, M. R. Delgado, et al., “Genetic algorithms to automatic weld bead detection in double wall double image digital radiographs,” in: J. Liu and C. Alippi (editors), Proc. of the WCCI 2012 IEEE World Congr. on Computational Intelligence (June 10–15, 2012, Brisbane, Australia), Springer (2012), pp. 1–7.Google Scholar
  4. 4.
    T. Mandziy, “Active shape models with adaptive weights,” Vidbir Obrob. Inform., Issue 30 (106), 133–137 (2009)Google Scholar
  5. 5.
    T. F. Cootes, C. J. Taylor, D. H. Cooper, and J. Graham, “Active shape models – their training and application,” Comput. Vision Image Understand., 61(1), 38–59 (1995).CrossRefGoogle Scholar
  6. 6.
    V. R. Rathod and R. S. Anand, “A novel method for detection and quantification of incomplete penetration type flaws in weldments,” J. X-Ray Sci. Technol., 19, 261–274 (2011).Google Scholar
  7. 7.
    R. A. Vorobel’, “Logarithmic processing of images. Part 1: Basic model,” Vidbir Obrob. Inform., Issue 31 (107), 26–35 (2009)Google Scholar
  8. 8.
    R. A. Vorobel’, Logarithmic Processing of Images [in Ukrainian], Naukova Dumka, Kyiv (2012).Google Scholar
  9. 9.
    V. V. Botsyan and R. A. Vorobel’, “Improvement of images with the help of an adaptive logarithmic-type model,” Vidbir Obrob. Inform., Issue 37(113), 86–92 (2012).Google Scholar
  10. 10.
    I. B. Ivasenko, “Localization and segmentation of objects on different scales with the use of information function,” Vidbir Obrob. Inform., Issue 22(98), 76–80 (2005).Google Scholar
  11. 11.
    R. A. Vorobel’, N. V. Opyr, Z. А. Bernyk, and O. R. Beregulyak, “Computer technology for the determination of sensitivity of X-ray testing according to an image of a grooved standard,” Defektoskopiya, No. 5, 81–89 (2009)Google Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • R. A. Vorobel
    • 1
  • I. B. Ivasenko
    • 1
  • T. S. Mandzii
    • 1
  • V. V. Botsyan
    • 1
  1. 1.Karpenko Physicomechanical InstituteUkrainian National Academy of SciencesLvivUkraine

Personalised recommendations