Journal of Nondestructive Evaluation

, Volume 25, Issue 2, pp 53–66 | Cite as

Neural Network Based Thickness Estimation from Multiple Radiographic Images

  • I. N. Tansel
  • F. Inanc
  • N. Reen
  • P. Chen
  • X. Wang
  • C. Kropas-Hughes
  • A. Yenilmez


Back propagation (BP) type artificial neural networks (ANN) have been trained and used for thickness estimations from radiographic images. Test objects have been assembled from different materials and radiographic images of the test objects were obtained for thickness estimations. While some of the study has been based on the synthetic images formed through the radiographic simulation program XRSIM, the rest of the study has used actual radiographic images. The average estimation errors were 7% and 9% when two and three synthetic radiographic images obtained at different x-ray tube settings were used. With the actual images, the thickness of only one of the materials has been estimated and the material was identified. This has been due to the fact that scattering of x-rays by the test object results in a non uniform gray scale variation in the radiographic images even though the object thickness is uniform.


Radiography radiographic simulation neural networks thickness estimation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    F. Inanc and J. N. Gray, A CAD interfaced simulation tool for X-ray NDE studies, in Review of Progress in QNDE, edited by D. O. Thomson and D. E. Chimenti, Vol. 9, Plenum, New York.Google Scholar
  2. 2.
    F. Inanc and J. N. Gray, Scattering Simulations in Radio- graphy, Appl. Radiat. Isot. 48(10–12), pp. 1299–1305 (1997).CrossRefGoogle Scholar
  3. 3.
    I. N. Tansel, N. Reen, and C. V. Kropas-Hughes, Neural network based material identification and part thickness estimation from two radiographic images, Artificial Neural Networks and Neural Information Processing—ICANN/ICONIP2003, Lecture Notes in Computer Science, 2003 Springer, pp. 1018–1025.Google Scholar
  4. 4.
    G. Berodias and M. G. Peix, Nondestructive measurement of density and effective atomic number by photon scattering, Mat. Eval. 46, pp. 1209–1213 (1988).Google Scholar
  5. 5.
    P. Engler and W. D. Friedman, Review of dual-energy computed tomography techniques, Mat. Eval. 48, pp. 623–629 (1990).Google Scholar
  6. 6.
    C. Robert-Coutant, V. Moulin, R. Sauze, P. Rizo, and J. M. Casagrande, Estimation of the matrix attenuation in heterogeneous radioactive waste drums using dual-energy computed tomography, Nucl. Instrum. Methods Phy. Res. A 422, pp. 949–956 (1999).CrossRefGoogle Scholar
  7. 7.
    R. Wojcik, S. Majewski, F. R. Parker, and W. P. Winfree, Single shot dual energy reverse geometry X-radiography, Nuclear Science Symposium. IEEE Meeting: November 2, 1996, pp. 811–815.Google Scholar
  8. 8.
    D. E. Rumelhart, G. Hilton, and R. J. Williams, Learning internal representations by error propagation, parallel distributed processing, in Explorations in the Microstructure of Cognition, Vol. 1, edited by E. Rumelhart and J. L. McClelland, MIT Press, pp. 319–362 (1986).Google Scholar
  9. 9.
    I. N. Tansel, W. Y. Bao, B. Tansel, and C. M. Jordahl, Modeling, contamination with trainable networks, fuzzy logic and evol. programming, in Intelligent Engineering Systems Through Artificial Neural Networks, Vol. 5, edited by CH. Dagli, M. Aksoy, C. L. Philip Chen, B. R. Fernandez, and J. Ghosh, ASME Press, pp. 823–828 (1995).Google Scholar
  10. 10.
    Casasent, David, Chen, Xue-Wen, New training strategies for RBF neural networks for X-ray agricultural product inspection, Pattern Recognit. 36(2), pp. 535–547 (2003).CrossRefGoogle Scholar
  11. 11.
    L. Bocchi, G. Coppini, R. De Dominicis, and G. Valli, Tissue characterization from X-ray images, Med. Eng. Phys. 19(4), pp. 336–342 (1997).CrossRefGoogle Scholar
  12. 12.
    J. K. Kim, J. M. Park, K. S. Song, and H. W. Park, Texture analysis and artificial neural network for detection of clustered micro calcifications on mammograms, Neural Networks for Signal Processing—Proceedings of the IEEE Workshop, pp. 199–206 (1997).Google Scholar
  13. 13.
    Neural Networks, Statsoft Electronic Textbook, http://www.
  14. 14.
    I. N. Tansel, B. Ozcelik, W. Y. Bao, P. Chen, D. Rincon, S. Y. Yang, and A. Yenilmez, Selection of Optimal Cutting Conditions By Using Gonns, Int. J. Machine Tools Manufacture 46(1), pp. 26–35 (2006). CrossRefGoogle Scholar

Copyright information

© Springer Science + Business Media, Inc. 2006

Authors and Affiliations

  • I. N. Tansel
    • 1
  • F. Inanc
    • 2
  • N. Reen
    • 1
  • P. Chen
    • 1
  • X. Wang
    • 1
  • C. Kropas-Hughes
    • 3
  • A. Yenilmez
    • 4
  1. 1.Florida International UniversityMiamiUSA
  2. 2.CNDEIowa State UniversityAmesUSA
  3. 3.AFRL/MLLP, Wright PattersonOHUSA
  4. 4.Technical University of IstanbulIstanbulTurkey

Personalised recommendations