Abstract
In the current NDT 4.0 revolution, machine learning and artificial intelligence have emerged as the major enablers for non-destructive testing and evaluation (NDT&E) of industrial components. However, recent developments in active thermal NDT (TNDT) support its use as a practical method for checking a range of industrial components. Additionally, recent post-processing research in TNDT has developed several machine learning models to replace human interaction and offer automatic defect detection. However, the smaller area of the flaws and their related few thermal profiles than the wide sound area, leading to imbalanced datasets, make it difficult to train a supervised deep neural. Recently added to TNDT are anomaly detection models and one-class classifiers, both of which are commonly applied machine learning models to real-world issues. The accuracy and other important metrics in autonomous defect detection are influenced by the hyper-parameters of these models, such as contamination factor, volume of training data, and initialization parameter of the relevant model. The current paper investigates how initialization parameters affect these models' TNDT capabilities for automated flaw detection. Using quadratic frequency modulated thermal wave imaging (QFMTWI), a carbon fiber-reinforced polymer specimen with variously sized artificially produced back-holes at different depths is examined. A good hyper-parameter for automatic flaw identification is chosen after qualitatively comparing testing accuracy, precision, recall, F-score, and probability.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig8_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig9_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig10_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig11_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig12_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS1061830923600600/MediaObjects/11181_2024_1507_Fig13_HTML.png)
REFERENCES
Aldrin, C. John, Intelligence augmentation and Human Machine Interface (HMI) best practices for NDT 4.0 reliability, ASNT Annu. Conf. (Las Vegas, 2019).
Maldague, X.P.V., Theory and Practice of Infrared Technology for Nondestructive Testing, New York: Wiley, 2001.
Ciampa Francesco, Pooya Mahmoodi, Fulvio Pinto, and Michele Meo, Recent advances in active infrared thermography for nondestructive testing of aerospace components, Sensors, 2018, vol. 18, no. 2, p. 609.
He Yunze, Baoyuan Deng, Hong** Wang, Liang Cheng, Ke Zhou, Siyuan Cai, and Francesco Ciampa, Infrared machine vision and infrared thermography with deep learning: A review, Infrared Phys. & Technol., 2021, vol. 116, p. 103754.
Luo, Q., Gao, B., Woo, W.L., and Yang, Y., Temporal and spatial deep learning network for infrared thermal defect detection, NDT & E Int., 2019, vol. 108, p. 102164.
Hu Bozhen, Bin Gao, Wai Lok Woo, Lingfeng Ruan, Jikun **, Yang Yang, and Yongjie Yu, A lightweight spatial and temporal multi-feature fusion network for defect detection, IEEE Trans. Imag. Process., 2020, vol. 30, pp. 472–486.
Saeed Numan, Nelson King, Zafar Said, and Mohammed A. Omar, Automatic defects detection in CFRP thermograms, using convolutional neural networks and transfer learning, Infrared Phys. & Technol., 2019, vol. 102, p. 103048.
Wei Ziang, Henrique Fernandes, Hans-Georg Herrmann, Jose Ricardo Tarpani, and Ahmad Osman, A deep learning method for the impact damage segmentation of curve-shaped CFRP specimens inspected by infrared thermography, Sensors, 2021, vol. 21, no. 2, p. 395.
Fang Qiang, Clemente Ibarra-Castanedo, and Xavier Maldague, Automatic defects segmentation and identification by deep learning algorithm with pulsed thermography: Synthetic and experimental data, Big Data Cognit. Comput., 2021, vol. 5, no. 1, p. 9.
Wei Ziang, Henrique Fernandes, Jose Ricardo Tarpani, Ahmad Osman, and Xavier Maldague, Stacked denoising autoencoder for infrared thermography image enhancement, 2021 IEEE 19th Int. Conf. Ind. Informatics (INDIN) (Palma de Mallorca, 2021), pp. 1–7.
Cheng Liangliang and Mathias Kersemans, Dual-IRT-GAN: A defect-aware deep adversarial network to perform super-resolution tasks in infrared thermographic inspection, Compos. B Eng., 2022, p. 110309.
Tretout, H., David, D., Marin, J.Y., Dessendre, M., Couet, M., and Avenas-Payan, I., An evaluation of artificial neural networks applied to infrared thermography inspection of composite aerospace structures, NDT & E Int., 1996, vol. 6, no. 29, p. 392.
Vijaya Lakshmi, A., Gopi tilak, V., Muzammil M. Parvez, Subhani, S.K., and Ghali, V.S., Artificial neural networks based quantitative evaluation of subsurface anomalies in quadratic frequency modulated thermal wave imaging, Infrared Phys. Technol., 2019, vol. 97, pp. 108–115.
Lakshmi A. Vijaya, Ghali, V.S., Subhani, S.K., and Naik R. Baloji, Automated quantitative subsurface evaluation of fiber reinforced polymers, Infrared Phys. & Technol., 2020, vol. 110, p. 103456.
Vesala, G.T., Ghali, V.S., Vijaya Lakshmi, A., and Naik, R.B., Deep and handcrafted feature fusion for automatic defect detection in quadratic frequency modulated thermal wave imaging, Russ. J. Nondestr. Test., 2021, vol. 57, pp. 476–485.
Liu Lishuai, Chenjun Guo, Yanxun **ang, Yanxin Tu, Liming Wang, and Fu-Zhen Xuan, A semi-supervised learning framework for recognition and classification of defects in transient thermography detection, IEEE Trans. Ind. Informatics, 2021, vol. 18, no. 4, pp. 2632–2640.
Morelli Davide, Roberto Marani, Ester D’Accardi, Davide Palumbo, Umberto Galietti, and Tiziana D’Orazio, A convolution residual network for heating-invariant defect segmentation in composite materials inspected by lock-in thermography, IEEE Trans. Instrum. Meas., 2021, vol. 70, pp. 1–14.
Tilak V. Gopi, Ghali, V.S., Vijaya Lakshmi, A., Suresh, B., and Naik, R.B., Proximity based automatic defect detection in quadratic frequency modulated thermal wave imaging, Infrared Phys. & Technol., 2021, vol. 114, p. 103674.
Vesala, G.T., Ghali, V.S., Rama Sastry, D.V.A., and Naik, R.B., Deep anomaly detection model for composite inspection in quadratic frequency modulated thermal wave imaging, NDT & E Int., 2022, vol. 132, p. 102710.
Munir Mohsin, Muhammad Ali Chattha, Andreas Dengel, and Sheraz Ahmed, A comparative analysis of traditional and deep learning-based anomaly detection methods for streaming data, 2019 18th IEEE Int. Conf. Mach. Learn. Appl. (ICMLA) (Boca Raton, 2019), pp. 561–566.
Schölkopf, B., Platt, J.C., Shawe-Taylor, J., Smola, A.J., and Williamson, R.C., Estimating the support of a high-dimensional distribution, Neural Comput., 2001, vol. 13, no. 7, pp. 1443–1471.
Liu Fei Tony, Kai Ming Ting, and Zhi-Hua Zhou, Isolation forest, 2008 8th IEEE Int. Conf. Data Min. (Pisa, 2008), pp. 413–422.
Breunig Markus M., Hans-Peter Kriegel, Raymond T. Ng, and Jörg Sander, LOF: identifying density-based local outliers, Proc. 2000 ACM SIGMOD Int. Conf. Manag. Data (Dallas, 2000), pp. 93–104.
Funding
This work was supported by ongoing institutional funding. No additional grants to carry out or direct this particular research were obtained.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
CONFLICT OF INTEREST
The authors of this work declare that they have no conflicts of interest.
ABBREVIATIONS
NDT&E | Non-destructive testing and evaluation |
TNDT | Thermal non-destructive testing |
QFMTWI | Quadratic frequency modulated thermal wave imaging |
ML | Machine learning |
DL | Deep learning |
CNN | Convolution neural network |
LSTM | Long-short-term memory |
RCNN | Region based convolution neural network |
YOLO | You only look once |
ADM | Anomaly detection model |
CFRP | Carbon fiber reinforced polymer |
OCSVM | One class support vector machine |
IF | Isolation forest |
LOF | Local outlier factor |
TP | True positive |
TN | True negative |
FP | False positive |
FN | False negative |
POD | Probability of detection |
SNR | Signal-to-noise ratio |
Additional information
Publisher’s Note.
Pleiades Publishing remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Vesala, G.T., Ghali, V.S., Prasanthi, Y.N. et al. Parametric Study of Anomaly Detection Models for Defect Detection in Infrared Thermography. Russ J Nondestruct Test 59, 1259–1271 (2023). https://doi.org/10.1134/S1061830923600600
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1061830923600600