Composite Material Defect Segmentation Using Deep Learning Models and Infrared Thermography

Iago Garcia Vargas; Henrique Fernandes

doi:10.22456/2175-2745.143066

Authors

Iago Garcia Vargas Universidade Federal de Uberlândia (UFU) https://orcid.org/0000-0001-9723-0571
Henrique Fernandes Universidade Federal de Uberlândia (UFU) & Cranfield University (CU) https://orcid.org/0000-0002-7078-9620

DOI:

https://doi.org/10.22456/2175-2745.143066

Keywords:

Infrared Thermography, Segmentation, Deep Learning, Composite Materials

Abstract

For non-destructive assessment, the segmentation of infrared thermographic images of carbon fiber composites is a critical task in material characterization and quality assessment. This study focuses on applying image processing techniques, particularly adaptive thresholding, alongside neural network models such as U-Net and DeepLabv3 for infrared image segmentation tasks. An experimental analysis was conducted on these networks to compare their performance in segmenting artificial defects from infrared images of a carbon-fibre reinforced polymer sample. The performance of these models was evaluated based on the F1-Score and Intersection over Union (IoU) metrics. The findings reveal that DeepLabv3 demonstrates superior results and efficiency in segmenting patterns of infrared images, achieving an F1-Score of 0.94 and an IoU of 0.74, showcasing its potential for advanced material analysis and quality control.

Downloads

Download data is not yet available.

References

MALDAGUE, X. Theory and practice of infrared technology for nondestructive testing. 2001.

MEOLA, C. Origin and theory of infrared thermography. In: Infrared Thermography Recent Advances and Future Trends. [S.l.]: Bentham eBooks, 2012. p. 3–28.

DONG, S.; WANG, P.; ABBAS, K. A survey on deep learning and its applications. Computer Science Review, v. 40, p. 100379, 2021.

KYLILI, A. et al. Infrared thermography (IRT) applications for building diagnostics: A review. Applied Energy, Elsevier, v. 134, p. 531–549, 2014.

HEGDE, S.; SHENOY, B. S.; CHETHAN, K. Review on carbon fiber reinforced polymer (CFRP) and their mechanical performance. Materials Today: Proceedings, Elsevier, v. 19, p. 658–662, 2019.

HEUER, H. et al. Review on quality assurance along the CFRP value chain – non-destructive testing of fabrics, preforms and CFRP by HF radio wave techniques. Composites Part B: Engineering, Elsevier, v. 77, p. 494–501, 2015.

SWIDERSKI, W. Non-destructive testing of CFRP by laser excited thermography. Composite Structures, Elsevier, v. 209, p. 710–714, 2019.

GUSTAFSSON, F.; GUSTAFSSON, F. Adaptive filtering and change detection. [S.l.]: Wiley New York, 2000. v. 1.

RAJINIKANTH, V.; RAJA, N. S. M.; DEY, N. A Beginner’s Guide to Multilevel Image Thresholding. [S.l.]: CRC Press, 2020.

SPEAKMAN, J. R.; WARD, S. Infrared thermography: principles and applications. Zoology-Jena-, GUSTAV FISCHER, v. 101, p. 224–232, 1998.

HE, Y. et al. Infrared machine vision and infrared thermography with deep learning: A review. Infrared Physics & Technology, Elsevier, v. 116, p. 103754, 2021.

RONNEBERGER, O.; FISCHER, P.; BROX, T. U-net: Convolutional networks for biomedical image segmentation. In: SPRINGER. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18. [S.l.], 2015. p. 234–241.

DING, Y. et al. A stacked multi-connection simple reducing net for brain tumor segmentation. IEEE Access, IEEE, v. 7, p. 104011–104024, 2019.

RAVIKUMAR, A.; SRIRAMAN, H. Acceleration of image processing and computer vision algorithms. In: Handbook of Research on Computer Vision and Image Processing in the Deep Learning Era. [S.l.]: IGI Global, 2023. p. 1–18.

CHEN, L.-C. et al. Rethinking atrous convolution for semantic image segmentation. arXiv preprint arXiv:1706.05587, 2017.

PEDRAYES, O. D. et al. Semantic segmentation for non-destructive testing with step-heating thermography for composite laminates. Measurement, Elsevier, v. 200, p. 111653, 2022.

SONAVANE, R. et al. Exploring ResNet101, InceptionV3, and Xception for Modi script character classification. International Journal of Intelligent Systems and Applications in Engineering, v. 12, n. 17s, p. 117–124, 2024.

FERNANDES, H. et al. Carbon fiber composites inspection and defect characterization using active infrared thermography: numerical simulations and experimental results. Applied Optics, Optica Publishing Group, v. 55, n. 34, p. D46–D53, 2016.

OBI, J. C. A comparative study of several classification metrics and their performances on data. World Journal of Advanced Engineering Technology and Sciences, v. 8, n. 1, p. 308–314, 2023.

RAINIO, O.; TEUHO, J.; KLÉN, R. Evaluation metrics and statistical tests for machine learning. Scientific Reports, Nature Publishing Group UK London, v. 14, n. 1, p. 6086, 2024.

PEDRAYES, O. D. et al. Semantic segmentation for non-destructive testing with step-heating thermography for composite laminates. Measurement, Elsevier, v. 200, p. 111653, 2022.

WANG, Y. et al. Digital image correlation (DIC) based damage detection for CFRP laminates by using machine learning based image semantic segmentation. International Journal of Mechanical Sciences, Elsevier, v. 230, p. 107529, 2022.