Vol. 20 No. 1 (2023): Journal of Non Destructive Testing and Evaluation (JNDE), March 2023
Research Papers

Neural Network Based Automatic Defect Detection in Infrared Thermography

Geetika Dua
Thapar Institute of Engineering and Technology, Patiala.147004
Pranav Yogesh Mahajan
Thapar Institute of Engineering and Technology, Patiala.147004
Aneesh Bajwa
Thapar Institute of Engineering and Technology, Patiala.147004
Vanita Arora
Indian Institute of Information Technology Una, Vill. Saloh, Teh. Haroli, Distt. Una Himachal Pradesh, India-177209

Published 12-03-2023


  • Defect detection,
  • Gabor wavelet,
  • Gabor Filter,
  • feed forward neural network classifier

How to Cite

Dua, G., Mahajan, P. Y., Bajwa, A., & Arora, V. (2023). Neural Network Based Automatic Defect Detection in Infrared Thermography. Journal of Non-Destructive Testing and Evaluation (JNDE), 20(1), 58–63. Retrieved from https://jnde.isnt.in/index.php/JNDE/article/view/33


Early, accurate, and automatic detection of defects is an essential aspect of quality improvement. This paper employs a classification-based automatic defect detection method using Gabor filter features. The thermal patterns of various defects are used to provide the primary information for defect detection. Considering the desirable characteristics of spatial locality and orientation selectivities of the Gabor filter, we design filters for extracting defect features from the thermogram. The feature vector based on Gabor filters is used as the classifier's input, a Feed forward neural network (FFNN) on a reduced feature subspace. A finite element method and experiment were adopted to simulate a Carbon fiber reinforced polymer (CFRP) material with void holes as defects. The thermogram will be convolved with Gabor filters by multiplying the image by Gabor filters in the frequency domain. Features are a cell array containing the result of the convolution of the image with each of the forty Gabor filters. The input vector of the network will have large values, which means a large amount of computation. So we reduce the matrix size to one-third of its original size by deleting some rows and columns. This work aims to implement a classifier based on neural networks (Multi-layer Perceptron) to differentiate defect and non-defect patterns.


  1. J. Tou, R. Gonzalez, Pattern Recognition Principles. Massachusetts, USA: Addison-Wesley Publishing Company, 1981.
  2. X. P. V. Maldague (2001), “Theory and Practice of Infrared Thermography for Non- Destructive Testing,” John Wiley & Sons Inc.
  3. D. P. Almond and S. K. Lau (1994), “Defect sizing by transient thermography I: An analytical treatment,” Journal of Physics D: Applied Physics, 27 (5), pp. 1063-1069.
  4. D. L. Balageas, J.C. Krapez, P. Cielo (1986), “Pulsed photothermal modeling of layered materials,” Journal of Applied Physics, 59 (2), pp. 348-357.
  5. X. Maldague and S. Marinetti (2003), “Pulse Phase Infrared Thermography,” Review of Scientific Instruments, 74 (1 II), pp. 417-419.
  6. X. Maldague and S. Marinetti (1996), “Pulse phase infrared thermography,” Journal of applied physics, vol. 79, no. 5, pp. 2694-2698.
  7. G. Busse (1979), “Optoacoustic phase angle measurement for probing a metal,” Applied Physics Letters, 35 (10), pp. 759-760.
  8. G. Busse, D. Wu, W. Karpen (1992), “Thermal wave imaging with phase-sensitive modulated thermography,” Journal of Applied Physics, 71 (8), pp. 3962-3965.
  9. R. Mulaveesala and S. Tuli (2006), “Theory of frequency modulated thermal wave imaging for nondestructive subsurface defect detection,” Applied Physics Letters, 89 (19), art.no. 91913.
  10. S. Tuli and R. Mulaveesala (2005), "Defect detection by pulse compression in frequency modulated thermal wave imaging," Quantitative InfraRed Thermography Journal, vol. 2(1), pp. 41-54.
  11. Ghali V S and Mulaveesala R (2010). “Frequency modulated thermal wave imaging techniques for non-destructive testing,” Insight: Non-Destructive Testing and Condition Monitoring52, 475-480.
  12. Arora V, Mulaveesala R and Bison P (2016), “Effect of Spectral Reshaping on Frequency Modulated Thermal Wave Imaging for Non-destructive Testing and Evaluation of Steel Material,” Journal of Nondestructive Evaluation 35, 1-7.
  13. Yang F and Paindavoine M (2003). “Prefiltering for pattern Recognition Using Wavelet Transform and Neural Networks,” Adavances in imaging and Electron physics, Vol. 127.
  14. Fang Q, Nguyen B D, Castanedo C I, Duan Y and Maldague II X (2020), “Automatic defect detection in infrared thermography by deep learning algorithm”, Thermosense: thermal infrared applications XLII SPIE) pp 180-195.
  15. Mulaveesala R and Dua G (2016), “Non-invasive and non-ionizing depth resolved infra-red imaging for detection and evaluation of breast cancer: A numerical study”, Biomedical Physics and Engineering Express, 2, 055004.
  16. Dua G, Arora V and Mulaveesala R (2021), “Defect Detection Capabilities of Pulse Compression Based Infrared Non-Destructive Testing and Evaluation”, IEEE Sensors Journal21, 7940-7947.
  17. Arora V, Mulaveesala R, Rani A, Kumar S, Kher V, Mishra P, Kaur J, Dua G and Jha R (2021), “Infrared Image Correlation for Non-destructive Testing and Evaluation of Materials”, Journal of Nondestructive Evaluation, 40, 1-7.
  18. Lyons M J, Budynek J and Akamatsu S (1999), “Automatic Classification of Single Facial Images”, IEEE Trans. on PAMI, Vol. 21(12).
  19. Kanade T (1973), “Picture processing by computer complex and recognition of human faces”, Technical report, Kyoto University, Dept. of Information Science.
  20. Turk M and Pentland A (1991), “Eigenfaces for recognition”, Journal of Cognitive Science, pp.71-86.