Multi-Scale Time Series Segmentation Network Based on Eddy Current Testing for Detecting Surface Metal Defects

Xiaorui Li; Xiaojuan Ban; Haoran Qiao; Zhaolin Yuan; Hong-Ning Dai; Chao Yao; Yu Guo; Mohammad S. Obaidat; George Q. Huang

IEEE/CAA Journal of Automatica Sinica

JCR Impact Factor: 15.3, Top 1 (SCI Q1)

CiteScore: 23.5, Top 2% (Q1)
Google Scholar h5-index: 77， TOP 5

Article Contents

Article Navigation > IEEE/CAA Journal of Automatica Sinica > 2025 > In Press, Accepted Manuscript

X. Li, X. Ban, H. Qiao, Z. Yuan, H.-N. Dai, C. Yao, Y. Guo, Mohammad S. Obaidat, and George Q. Huang, “Multi-scale time series segmentation network based on Eddy current testing for detecting surface metal defects,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 3, pp. 1–11, Mar. 2025.

Citation:

X. Li, X. Ban, H. Qiao, Z. Yuan, H.-N. Dai, C. Yao, Y. Guo, Mohammad S. Obaidat, and George Q. Huang, “Multi-scale time series segmentation network based on Eddy current testing for detecting surface metal defects,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 3, pp. 1–11, Mar. 2025.

Citation:

X. Li, X. Ban, H. Qiao, Z. Yuan, H.-N. Dai, C. Yao, Y. Guo, Mohammad S. Obaidat, and George Q. Huang, “Multi-scale time series segmentation network based on Eddy current testing for detecting surface metal defects,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 3, pp. 1–11, Mar. 2025.

PDF( 2857 KB)

Multi-Scale Time Series Segmentation Network Based on Eddy Current Testing for Detecting Surface Metal Defects

Funds: This work was supported by the National Science and Technology Major Project of the Ministry of Science and Technology of China (2024ZD0608100) and the National Natural Science Foundation of China (62332017, U22A2022)

More Information

Author Bio:
Xiaorui Li received the B.S. degree in computer science from University of Science and Technology in 2020. He is a Ph.D. candidate in computer science and technology at the School of Computer and Communication Engineering, University of Science and Technology Beijing. His research interests covers time series forecasting, operational state detection and process control

Xiaojuan Ban received the Ph.D. degree from the University of Science and Technology Beijing in 2003. She is currently a Ph.D. Supervisor with the University of Science and Technology Beijing. She is also the Managing Director of the Chinese Association for Artificial Intelligence. Her research interests include artificial intelligence, natural human-computer interaction and 3D visualization technology. She has authored more than 300 articles. She is also a Member of the standing committee of the human-computer Interaction specialty and the theoretical computer science specialty in the China Computer Federation

Haoran Qiao received the B.S. degree in computer science from University of Science and Technology in 2022. He is now a master student in the School of Intelligence Science and Technology, University of Science and Technology Beijing. His current research interests complex industrial system, time series segmentation, time series classification

Zhaolin Yuan received the Ph.D. degree in computer science from the same university in 2023, followed by postdoctoral research at Huawei’s 2012 Central Research Institute and Hong Kong Polytechnic University. His work, supported by significant government and industry grants, focuses on artificial intelligence for mining and smart manufacturing. He is a Distinguished Associate Professor at the School of Intelligent Science and Technology, University of Science and Technology Beijing. His research interests include intelligent mining, intelligent manufacturing, LLM applications, industrial systems optimization, and reinforcement learning. He has published over 10 papers in top AI conferences (e.g., AAAI) and leading journals (e.g., IEEE TNNLS, IEEE TII)

Hong-Ning Dai (Senior Member, IEEE) received the Ph.D. degree in computer science and engineering from the Department of Computer Science and Engineering, the Chinese University of Hong Kong, China in 2008. He is currently with the Department of Computer Science at Hong Kong Baptist University, Hong Kong, China as an Associate Professor. His research interests include Internet of things, blockchain, big data analytics. He has served as Associate Editors/Editors for IEEE COMST, IEEE TII, and IEEE T-ITS

Chao Yao received the M.E. and the Ph.D. degrees in information security from Beijing Jiaotong University (BJTU) in 2010 and 2016, respectively. From 2014 to 2015, he was a visiting Ph.D. student with LTS4 Group, Institute of the Swiss Federal Institute of Technology (EPFL), Switzerland. He is currently an Associated Professor with the School of Computer and Communication Engineering, University of Science and Technology Beijing (USTB). His research interests include image/video compression, computer vision and human-computer interaction

Yu Guo received the B.S. degree in communication engineering and the Ph.D. degree in communication and information systems from the University of Science and Technology Beijing, in 2014 and 2020, respectively. He is currently a Lecturer with the Beijing Advanced Innovation Center for Materials Genome Engineering, University of Science and Technology Beijing. His main research interests include mobile wireless sensor and actor networks, cloud computing, and multirobot systems

Mohammad S. Obaidat (Life Fellow of IEEE) received the Ph.D. degree in computer engineering with a minor in computer science from The Ohio State University, USA. He is recognized all over the world for his pioneering and lasting contributions to several ICT areas. He is a Distinguished Professor at King Abdullah II School of Information Technology (KASIT), University of Jordan, Jordan; SRM University, India; University of Science and Technology, Beijing; and Amity University, India. He served as the Founding Dean of the College of Computing and Informatics, University of Sharjah, Dean of the College of Engineering at Prince Sultan University, and as the Advisor to the President of Philadelphia University. He served as a Professor and Chair of Departments at various US universities. He also served as the President of the Society for Modelling and Simulation International, SCS. His research interests include complex system modeling and simulation, green ICT (information and communications technology), AI and machine learning, cybersecurity, data analytics, and wireless networks. He published over 1200 refereed articles, 110 books and 70 book chapters. He has been ranked recently by ScholarGPS as the top Lifetime scholar in the World in Telecommunications Systems Management and the top scholar in Telecommunication in the World in prior 5 Years

George Q. Huang (Fellow of IEEE) received the B.Eng. degree in mechanical engineering from Southeast University in 1983, and the Ph.D. degree in mechanical engineering from Cardiff University, UK in 1991. He joined the Department of Industrial and Systems Engineering, Hong Kong Polytechnic University, Hong Kong, China in December 2022 as the Chair Professor of smart manufacturing. His main research interests include smart manufacturing and logistics, cyber-physical Internet,and systems analytics
Corresponding author: Xiaojuan Ban, e-mail: banxj@ustb.edu.cn
Received Date: 2024-07-22
Revised Date: 2024-10-15
Accepted Date: 2024-12-12

Available Online: 2025-01-09

Abstract

Abstract

In high-risk industrial environments like nuclear power plants, precise defect identification and localization are essential for maintaining production stability and safety. However, the complexity of such a harsh environment leads to significant variations in the shape and size of the defects. To address this challenge, we propose the multivariate time series segmentation network (MSSN), which adopts a multiscale convolutional network with multi-stage and depth-separable convolutions for efficient feature extraction through variable-length templates. To tackle the classification difficulty caused by structural signal variance, MSSN employs logarithmic normalization to adjust instance distributions. Furthermore, it integrates classification with smoothing loss functions to accurately identify defect segments amid similar structural and defect signal subsequences. Our algorithm evaluated on both the Mackey-Glass dataset and industrial dataset achieves over 95% localization and demonstrates the capture capability on the synthetic dataset. In a nuclear plant’s heat transfer tube dataset, it captures 90% of defect instances with 75% middle localization F1 score.
- Eddy current testing,
- nondestructive testing,
- semantic segmentation,
- time series analysis

FullText(HTML)

References(30)

References

[1]	M. Pan, Y. He, G. Tian, D. Chen, and F. Luo, “Pec frequency band selection for locating defects in two-layer aircraft structures with air gap variations,” IEEE Trans. Instrumentation and Measurement, vol. 62, no. 10, pp. 2849–2856, 2013. doi: 10.1109/TIM.2013.2239892
[2]	A. N. AbdAlla, M. A. Faraj, F. Samsuri, D. Rifai, K. Ali, and Y. AlDouri, “Challenges in improving the performance of Eddy current testing,” Measurement and Control, vol. 52, no. 1–2, pp. 46–64, 2019. doi: 10.1177/0020294018801382
[3]	X. Huang, W. Qu, and L. Xiao, “Identification method of internal leakage in nuclear power plants valves using convolutional block attention module,” Nuclear Engineering and Design, vol. 424, p. 113239, 2024.
[4]	G. D’Angelo, M. Laracca, S. Rampone, and G. Betta, “Fast Eddy current testing defect classification using Lissajous figures,” IEEE Trans. Instrumentation and Measurement, vol. 67, no. 4, pp. 821–830, 2018. doi: 10.1109/TIM.2018.2792848
[5]	Y. Tao, H. Xu, Z. Chen, R. Huang, Q. Ran, Q. Zhao, H. Yan, Z. Zhang, and W. Yin, “Automatic feature extraction method for crack detection in eddy current testing,” in Proc. IEEE Int. Instrumentation and Measurement Technology Conf., 2019, pp. 1–6.
[6]	H. Phan, K. Mikkelsen, O. Y. Chén, P. Koch, A. Mertins, and M. De Vos, “x “Sleeptransformer: Automatic sleep staging with interpretability and uncertainty quantification,” IEEE Trans. Biomedical Engineering, vol. 69, no. 8, pp. 2456–2467, 2022. doi: 10.1109/TBME.2022.3147187
[7]	G. Li, W. Yan, and Z. Wu, “Discovering shapelets with key points in time series classification,” Expert Systems Applications, vol. 132, pp. 76–86, 2019. doi: 10.1016/j.eswa.2019.04.062
[8]	T. A. Alvarenga, A. L. Carvalho, L. M. Honorio, A. S. Cerqueira, L. M. Filho, and R. A. Nobrega, “Detection and classification system for rail surface defects based on Eddy current,” Sensors, vol. 21, p. 23, 2021.
[9]	T. Meng, Y. Tao, Z. Chen, et al., “Depth evaluation for metal surface defects by Eddy current testing using deep residual convolutional neural networks,” IEEE Trans. Instrumentation and Measurement, vol. 70, pp. 1–13, 2021.
[10]	X. Fu, C. Zhang, X. Peng, L. Jian, and Z. Liu, “Towards end-toend pulsed Eddy current classification and regression with CNN,” in Proc. IEEE Int. Instrumentation and Measurement Technology Conf., 2019, pp. 1–5.
[11]	M. Afrasiabi, H. Khotanlou, and M. Mansoorizadeh, “DTW-CNN: Time series-based human interaction prediction in videos using CNN-extracted features,” The Visual Computer, vol. 36, pp. 1127–1139, 2020. doi: 10.1007/s00371-019-01722-6
[12]	M. Middlehurst, J. Large, M. Flynn, J. Lines, A. Bostrom, and A. Bagnall, “HIVE-COTE 2.0: A new meta ensemble for time series classification,” Machine Learning, vol. 110, no. 11–12, pp. 3211–3243, 2021. doi: 10.1007/s10994-021-06057-9
[13]	H. Ismail Fawaz, B. Lucas, G. Forestier, C. Pelletier, D. F. Schmidt, J. Weber, G. I. Webb, L. Idoumghar, P.-A. Muller, and F. Petitjean, “Inceptiontime: Finding alexnet for time series classification,” Data Mining and Knowledge Discovery, vol. 34, no. 6, pp. 1936–1962, 2020. doi: 10.1007/s10618-020-00710-y
[14]	Z. Yuan, Y. Wang, X. Ban, C. Ning, H.-N. Dai, and H. Wang, “Autonomous-jump-ODENet: Identifying continuous-time jump systems for cooling-system prediction,” IEEE Trans. Industrial Inform., vol. 19, no. 7, pp. 7894–7904, 2023. doi: 10.1109/TII.2022.3207835
[15]	M. Perslev, S. Darkner, L. Kempfner, M. Nikolic, P. J. Jennum, and C. IGEL, “U-sleep: Resilient high-frequency sleep staging,” NPJ Digital Medicine, vol. 4, no. 1, p. 72, 2021. doi: 10.1038/s41746-021-00440-5
[16]	Z. Ebrahimi, M. Loni, M. Daneshtalab, and A. Gharehbaghi, “A review on deep learning methods for ECG arrhythmia classification,” Expert Systems Applications: X, vol. 7, p. 100033, 2020. doi: 10.1016/j.eswax.2020.100033
[17]	P. Liu, X. Sun, Y. Han, Z. He, W. Zhang, and C. Wu, “Arrhythmia classification of LSTM autoencoder based on time series anomaly detection,” Biomedical Signal Processing and Control, vol. 71, p. 103228, 2022. doi: 10.1016/j.bspc.2021.103228
[18]	S. Gaugel and M. Reichert, “PrecTime: A deep learning architecture for precise time series segmentation in industrial manufacturing operations,” Engineering Applications of Artificial Intelligence, vol. 122, p. 106078, 2023. doi: 10.1016/j.engappai.2023.106078
[19]	Z. Wang, L. Wang, C. Huang, Z. Zhang, and X. Luo, “Soil-moisturesensor-based automated soil water content cycle classification with a hybrid symbolic aggregate approximation algorithm,” IEEE Internet of Things J., vol. 8, no. 18, pp. 14003–14012, 2021. doi: 10.1109/JIOT.2021.3068379
[20]	Y. Wu, H.-N. Dai, and H. Tang, “Graph neural networks for anomaly detection in industrial internet of things,” IEEE Internet of Things J., vol. 9, no. 12, pp. 9214–9231, 2022. doi: 10.1109/JIOT.2021.3094295
[21]	S. Lu and S. Huang, “Segmentation of multivariate industrial time series data based on dynamic latent variable predictability,” IEEE Access, vol. 8, pp. 112092–112103, 2020. doi: 10.1109/ACCESS.2020.3002257
[22]	W. Wang, Q. Lin, D. Cai, and M. Li, “Similarity measurement of segment-level speaker embeddings in speaker diarization,” IEEE/ACM Trans. Audio, Speech, and Language Processing, vol. 30, pp. 2645–2658, 2022. doi: 10.1109/TASLP.2022.3196178
[23]	D. M. Sime, G. Wang, Z. Zeng, and B. Peng, “Deep learning-based automated steel surface defect segmentation: A comparative experimental study,” Multimedia Tools and Applications, vol. 83, no. 1, pp. 2995–3018, 2024. doi: 10.1007/s11042-023-15307-y
[24]	S. Li, Y. A. Farha, Y. Liu, M.-M. Cheng, and J. Gall, “MS-TCN++: Multistage temporal convolutional network for action segmentation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 45, no. 6, pp. 6647–6658, 2020.
[25]	S. Xie, Y. Xie, and T. Huang, “TSTFNN: Performance enhancement for fuzzy neural network in performance monitoring of industrial flotation processes,” IEEE Trans. Industrial Informatics, vol. 20, no. 3, pp. 4919–4929, 2024. doi: 10.1109/TII.2023.3330342
[26]	N. Tatbul, T. J. Lee, S. Zdonik, M. Alam, and J. Gottschlich, “Precision and recall for time series,” Advances in Neural Inform. Proce. Systems, vol. 31, pp. 1920–1930, 2018.
[27]	T.-D. Bui, V.-D. Pham, and T.-L. Cung, “Multilayer perceptron neural network and Eddy current technique for estimation of the crack depth on massive metal structures,” J. Military Science and Technology, vol. 77, pp. 3–12, 2022.
[28]	M. Middlehurst, P. Schäfer, and A. Bagnall, “Bake off redux: A review and experimental evaluation of recent time series classification algorithms,” Data Mining and Knowledge Discovery, vol. 38, pp. 1958–2031, 2024.
[29]	X. Ke, X. Miao, and W. Guo, “U-transformer-based multi-levels refinement for weakly supervised action segmentation,” Pattern Recognition, vol. 149, p. 110199, 2024. doi: 10.1016/j.patcog.2023.110199
[30]	M. Perslev, M. Jensen, S. Darkner, P. J. r. Jennum, and C. Igel, “U-time: A fully convolutional network for time series segmentation applied to sleep saging,” Advances in Neural Information Proce. Systems, vol. 32, pp. 4415–4426, 2019.

Supplements(0)

Cited By

Proportional views

Proportional views

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

Figures(9) / Tables(4)

Get Citation

PDF

XML

Article Metrics

Article views (54) PDF downloads(16)

Multi-Scale Time Series Segmentation Network Based on Eddy Current Testing for Detecting Surface Metal Defects

Abstract

References

Proportional views

Catalog

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Export File

Citation

Format

Content