Detecting Defects in Materials Using Deep Convolutional Neural Networks

  • Quentin BoyadjianEmail author
  • Nicolas Vanderesse
  • Matthew Toews
  • Philippe Bocher
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12131)


This paper proposes representing and detecting manufacturing defects at the micrometre scale using deep convolutional neural networks. The information theoretic notion of entropy is used to quantify the information gain or mutual information of filters throughout the network, where the deepest network layers are generally shown to exhibit the highest mutual information between filter responses and defects, and thus serve as the most discriminative features. Quantitative detection experiments based on the AlexNet architecture investigate a variety of design parameters pertaining to data preprocessing and network architecture, where the optimal architectures achieve an average accuracy of 98.54%. CNNs are relatively easy to perform and give impressive achievements in classification tasks. However, the informational complexity coming from the depth of networks represents a limit to improve their capabilities.


Convolutional neural network AlexNet Ti-6Al-4V Texture classification 


  1. 1.
    Gauthier, G., Coster, M., Chermant, L., Chermant, J.-L.: Morphological segmentation of cutting tools. Microsc. Microanal. Microstruct. 7(5–6), 339–344 (1996)CrossRefGoogle Scholar
  2. 2.
    Wejrzanowski, T., Spychalski, W., Różniatowski, K., Kurzydłowski, K.: Image based analysis of complex microstructures of engineering materials. Int. J. Appl. Math. Comput. Sci. 18(1), 33–39 (2008)CrossRefGoogle Scholar
  3. 3.
    Lee, S.G., Mao, Y., Gokhale, A.M., Harris, J., Horstemeyer, M.F.: Application of digital image processing for automatic detection and characterization of cracked constituent particles/inclusions in wrought aluminum alloys. Mater. Charact. 60(9), 964–970 (2009)CrossRefGoogle Scholar
  4. 4.
    Banerjee, S., Ghosh, S.K., Datta, S., Saha, S.K.: Segmentation of dual phase steel micrograph: an automated approach. Measurement 46(8), 2435–2440 (2013)Google Scholar
  5. 5.
    Vanderesse, N., Anderson, M., Bridier, F., Bocher, P.: Inter- and intragranular delta phase quantitative characterization in inconel 718 by means of image analysis. J. Microsc. 261(1), 79–87 (2016)CrossRefGoogle Scholar
  6. 6.
    Meimandi, S., Vanderesse, N., Thibault, D., Bocher, P., Viens, M.: Macro-defects characterization in cast CA-6NM martensitic stainless steel. Mater. Charact. 124, 31–39 (2017)CrossRefGoogle Scholar
  7. 7.
    Dutta, T., Das, D., Banerjee, S., Saha, S.K., Datta, S.: An automated morphological classification of ferrite-martensite dual-phase microstructures. Measurement 137, 595–603 (2019)Google Scholar
  8. 8.
    Cord, A., Bach, F., Jeulin, D.: Texture classification by statistical learning from morphological image processing: application to metallic surfaces. J. Microsc. 239, 159–166 (2010)Google Scholar
  9. 9.
    Ducato, A., Fratini, L., La Cascia, M., Mazzola, G.: An automated visual inspection system for the classification of the phases of Ti-6Al-4V titanium alloy. In: Wilson, R., Hancock, E., Bors, A., Smith, W. (eds.) CAIP 2013. LNCS, vol. 8048, pp. 362–369. Springer, Heidelberg (2013). Scholar
  10. 10.
    DeCost, B.L., Holm, E.A.: A computer vision approach for automated analysis and classification of microstructural image data. Comput. Mater. Sci. 110, 126–133 (2015)CrossRefGoogle Scholar
  11. 11.
    Jeulin, D.: Morphological probabilistic hierarchies for texture segmentation. Math. Morphol. - Theory Appl. 1(1) (2016)Google Scholar
  12. 12.
    Gupta, S., Sarkar, J., Kundu, M., Bandyopadhyay, N.R., Ganguly, S.: Automatic recognition of SEM microstructure and phases of steel using LBP and random decision forest operator. Measurement 151, 107224 (2020)Google Scholar
  13. 13.
    Chowdhury, A., Kautz, E., Yener, B., Lewis, D.: Image driven machine learning methods for microstructure recognition. Comput. Mater. Sci. 123, 176–187 (2016)CrossRefGoogle Scholar
  14. 14.
    DeCost, B.L., Francis, T., Holm, E.A.: Exploring the microstructure manifold: Image texture representations applied to ultrahigh carbon steel microstructures. Acta Materialia 133, 30–40 (2017)CrossRefGoogle Scholar
  15. 15.
    Lin, J., Ma, L., Yao, Y.: Segmentation of casting defect regions for the extraction of microstructural properties. Eng. Appl. Artif. Intell. 85, 150–163 (2019)CrossRefGoogle Scholar
  16. 16.
    Stan, T., Thompson, Z.T., Voorhees, P.W.: Optimizing convolutional neural networks to perform semantic segmentation on large materials imaging datasets: X-ray tomography and serial sectioning. Mater. Charact. 160, 110119 (2020)CrossRefGoogle Scholar
  17. 17.
    Dimiduk, D.M., Holm, E.A., Niezgoda, S.R.: Perspectives on the impact of machine learning, deep learning, and artificial intelligence on materials, processes, and structures engineering. Integr. Mater. Manuf. Innov. 7(3), 157–172 (2018)CrossRefGoogle Scholar
  18. 18.
    Vander Voort, G.F.: ASM Handbook, Volume 9: Metallography And Microstructures (ASM Handbook). ASM International (2004)Google Scholar
  19. 19.
    Lütjering, G., Williams, J.C.: Titanium. Springer, New York (2007).
  20. 20.
    Chrapoński, J., Szkliniarz, W.: Quantitative metallography of two-phase titanium alloys. Mater. Charact. 46(2–3), 149–154 (2001)CrossRefGoogle Scholar
  21. 21.
    Tiley, J., et al.: Quantification of microstructural features in \(\alpha \beta \) titanium alloys. Mater. Sci. Eng.: A 372(1–2), 191–198 (2004)Google Scholar
  22. 22.
    Vanderesse, N., Maire, E., Darrieulat, M., Montheillet, F., Moreaud, M., Jeulin, D.: Three-dimensional microtomographic study of widmanstätten microstructures in an alpha/beta titanium alloy. Scripta Materialia 58(6), 512–515 (2008)CrossRefGoogle Scholar
  23. 23.
    Li, H., Ji, Z., Yang, H.: Quantitative characterization of lamellar and equiaxed alpha phases of (\(\alpha + \beta \)) titanium alloy using a robust approach for touching features splitting. Mater. Charact. 76, 6–20 (2013)CrossRefGoogle Scholar
  24. 24.
    Zhao, H., Ho, A., Davis, A., Antonysamy, A., Prangnell, P.: Automated image mapping and quantification of microstructure heterogeneity in additive manufactured TI6AL4V. Mater. Charact. 147, 131–145 (2019)CrossRefGoogle Scholar
  25. 25.
    Sharma, H., van Bohemen, S.M.C., Petrov, R.H., Sietsma, J.: Three-dimensional analysis of microstructures in titanium. Acta Materialia 58(7), 2399–2407 (2010)Google Scholar
  26. 26.
    Foltz, J.W., Welk, B., Collins, P.C., Fraser, H.L., Williams, J.C.: Formation of grain boundary \(\alpha \) in \(\beta \) Ti alloys: its role in deformation and fracture behavior of these alloys. Metall. Mater. Trans. A: Phys. Metall. Mater. Sci. 42, 645–650 (2011)CrossRefGoogle Scholar
  27. 27.
    Chatfield, K., Simonyan, K., Vedaldi, A., Zisserman, A.: Return of the devil in the details: delving deep into convolutional nets (2014)Google Scholar
  28. 28.
    Liu, L., Chen, J., Fieguth, P., Zhao, G., Chellappa, R., Pietikäinen, M.: From BoW to CNN: two decades of texture representation for texture classification. Int. J. Comput. Vis. 127(1), 74–109 (2018)CrossRefGoogle Scholar
  29. 29.
    Liu, L., et al.: Deep learning for generic object detection: A survey. CoRR, abs/1809.02165 (2018)Google Scholar
  30. 30.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Pereira, F., Burges, C.J.C., Bottou, L., Weinberger, K.Q. (eds.) Advances in Neural Information Processing Systems 25, pp. 1097–1105. Curran Associates Inc. (2012)Google Scholar
  31. 31.
    Wang, S.H., Lv, Y.D., Sui, Y., Liu, S., Wang, S.J., Zhang, Y.D.: Alcoholism detection by data augmentation and convolutional neural network with stochastic pooling. J. Med. Syst. 42(1), 2 (2018)Google Scholar
  32. 32.
    Cover, T.M., Thomas, J.A.: Elements of Information Theory. Wiley, Hoboken (2012)Google Scholar
  33. 33.
    Tishby, N., Zaslavsky, N.: Deep learning and the information bottleneck principle. In: 2015 IEEE Information Theory Workshop (ITW), pp. 1–5. IEEE (2015)Google Scholar
  34. 34.
    Chaddad, A., Naisiri, B., Pedersoli, M., Granger, E., Desrosiers, C., Toews, M.: Modeling Information Flow Through Deep Neural Networks. arXiv e-prints, page arXiv:1712.00003, November 2017
  35. 35.
    Schneider, C.A., Rasband, W.S., Eliceiri, K.W.: NIH image to ImageJ: 25 years of image analysis. Nat. Methods 9(7), 671–675 (2012)Google Scholar
  36. 36.
    Schindelin, J., et al.: Fiji: an open-source platform for biological-image analysis. Nat. Methods 9(7), 676–682 (2012)Google Scholar
  37. 37.
    Abadi, M., et al.: TensorFlow: large-scale machine learning on heterogeneous systems (2015). Software available from tensorflow.orgGoogle Scholar
  38. 38.
    Chollet, F., et al.: Keras (2015).
  39. 39.
    Wei, J.-M., Yuan, X.-J., Qing-Hua, H., Wang, S.-Q.: A novel measure for evaluating classifiers. Expert Syst. Appl. 37(5), 3799–3809 (2010)CrossRefGoogle Scholar
  40. 40.
    Delgado, R., Núñez-González, J.D.: Enhancing confusion entropy as measure for evaluating classifiers. In: Graña, M., et al. (eds.) SOCO’18-CISIS’18-ICEUTE’18. AISC, vol. 771, pp. 79–89. Springer, Cham (2019). Scholar
  41. 41.
    Andrearczyk, V., Whelan, P.F.: Using filter banks in convolutional neural networks for texture classification. Pattern Recogn. Lett. 84, 63–69 (2016)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.École de Technologie SupérieureMontrealCanada

Personalised recommendations