journal article·2025-09-01

IDWS-Net A Robust Swin Transformer Architecture with Hierarchical Attention for Internal Weld Defect Segmentation

Tian Liang, Erqing Zhang, Shaofeng Wang, Ting Gao, Luncai Zhou, Wenjing Liu

Materials Evaluation· Volume 83· Issue 9
Publisher page

Abstract

Ultrasonic testing (UT) technology, owing to its advantages in defect detection efficiency and imaging quality, enables quantitative characterization of defects through ultrasonic images and plays a critical role in weld defect assessment. However, constrained by the presence of boundary blurring and high-intensity acoustic artifacts, existing segmentation methods struggle to meet the precision requirements necessary for quality analysis. To address these limitations, this study proposes a robust Swin Transformer architecture with hierarchical attention for internal weld defect segmentation (IDWS-Net). Two modules are introduced: the edge-aware enhancement module (EAEM), which improves boundary accuracy by hierarchically integrating multiscale edge features and contour attention mechanisms, and the multiscale context-aware gating (MCAG) module, which suppresses false defects caused by noise by utilizing global–local feature contrast. Experiments were conducted on 4000 collected and labeled B-scan images. Compared with other classic network structures, IDWS-Net demonstrated excellent segmentation performance across various defect types and sizes.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Welding Techniques and Residual Stresses · Non-Destructive Testing Techniques · Hydrogen embrittlement and corrosion behaviors in metals

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (28)

Show source-provided references
1. 1. Mondo, G., B. R. Moharana, S. A. Syed, and K. Muduli. 2024. “Evaluation of Weldability and Optimization of TIG Welding Process Parameters for Maximized Yield Strength.” In Recent Advancements in Product Design and Manufacturing Systems, eds. Deepak, B. B. V. L., M. V. A. Raju Bahubalendruni, D. R. K. Parhi, and B. B. Biswal, 135–146. Springer. https://doi.org/10.1007/978-981-97-6732-8_12.10.1007/978-981-97-6732-8_12 ↗
2. 2. Chen, Y., and Y. Wu. 2025. “Detection of Welding Defects Tracked by YOLOv4 Algorithm.” Applied Sciences (Basel, Switzerland) 15 (4): 2026. https://doi.org/10.3390/app15042026.10.3390/app15042026 ↗
3. 3. Madhvacharyula, A. S., A. V. S. Pavan, S. Gorthi, S. Chitral, N. Venkaiah, and D. V. Kiran. 2022. “In situ detection of welding defects: A review.” Welding in the World 66 (4): 611–28. https://doi.org/10.1007/s40194-021-01229-6.10.1007/s40194-021-01229-6 ↗
4. 4. Yang, D., H. Jiang, S. Ai, T. Yang, Z. Zhi, D. Jing, J. Gao, K. Yue, H. Cheng, and Y. Xu. 2024. “Detection method for weld defects in time-of-flight diffraction images based on multi-image fusion and feature hybrid enhancement.” Engineering Applications of Artificial Intelligence 138 (Part B): 109442. https://doi.org/10.1016/j.engappai.2024.109442.10.1016/j.engappai.2024.109442 ↗
5. 5. Lv, G., Z. Yao, D. Chen, Y. Li, H. Cao, A. Yin, Y. Liu, and S. Guo. 2023. “Fast and high-resolution laser-ultrasonic imaging for visualizing subsurface defects in additive manufacturing components.” Materials & Design 225: 111454. https://doi.org/10.1016/j.matdes.2022.111454.10.1016/j.matdes.2022.111454 ↗
6. 6. Taheri, H., and A. A. Hassen. 2019. “Nondestructive ultrasonic inspection of composite materials: A comparative advantage of phased array ultrasonic.” Applied Sciences (Basel, Switzerland) 9 (8): 1628. https://doi.org/10.3390/app9081628.10.3390/app9081628 ↗
7. 7. Bansal, A., S. C. Vettivel, M. Kumar, and M. Agarwal. 2023. “Weld defect identification and characterization in radiographic images using deep learning.” Engineering Research Express 5 (2): 025079. https://doi.org/10.1088/2631-8695/acdf3f.10.1088/2631-8695/acdf3f ↗
8. 8. Chen, C., S. Wang, and S. Huang. 2023. “An improved, faster RCNN-based weld ultrasonic atlas defect detection method.” Measurement and Control 56 (3-4): 832–43. https://doi.org/10.1177/00202940221092030.10.1177/00202940221092030 ↗
9. 9. Medak, D., L. Posilovic′, M. Subašic′, M. Budimir, and S. Loncaric. 2021. “Deep learning-based defect detection from sequences of ultrasonic B-scans.” IEEE Sensors Journal 22 (3): 2456–63. https://doi.org/10.1109/JSEN.2021.3134452.10.1109/jsen.2021.3134452 ↗
10. 10. S. A. Taghanaki, K. Abhishek, J. P. Cohen, J. Cohen-Adad, and G. Hamarneh. 2021. “Deep semantic segmentation of natural and medical images: A review.” Artificial Intelligence Review 54 (1): 137–78. https://doi.org/10.1007/s10462-020-09854-1.10.1007/s10462-020-09854-1 ↗
11. 11. Bang, H. T., S. Park, and H. Jeon. 2020. “Defect identification in composite materials via thermography and deep learning techniques.” Composite Structures 246: 112405. https://doi.org/10.1016/j.compstruct.2020.112405.10.1016/j.compstruct.2020.112405 ↗
12. 12. Xu, H., Z. H. Yan, B. W. Ji, P. F. Huang, J. P. Cheng, and X. D. Wu. 2022. “Defect detection in welding radiographic images based on semantic segmentation methods.” Measurement: Journal of the International Measurement Confederation 188: 110569. https://doi.org/10.1016/j.measurement.2021.110569.10.1016/j.measurement.2021.110569 ↗
13. 13. Ronneberger, O., P. Fischer, and T. Brox. 2015. “U-net: Convolutional networks for biomedical image segmentation.” In Medical image computing and computer-assisted intervention – MICCAI 2015, Lecture Notes in Computer Science 9351: 234–241, eds. Navab, N., J. Hornegger, W. M. Wells, and A. F. Frangi. Springer Cham. https://doi.org/10.1007/978-3-319-24574-4_28.10.1007/978-3-319-24574-4_28 ↗
14. 14. Sun, H., P. Ramuhalli, and R. E. Jacob. 2023. “Machine learning for ultrasonic nondestructive examination of welding defects: A systematic review.” Ultrasonics 127: 106854. https://doi.org/10.1016/j.ultras.2022.106854.10.1016/j.ultras.2022.106854 ↗
15. 15. Mohandas, R., P. Mongan, and M. Hayes. 2024. “Ultrasonic Weld Quality Inspection Involving Strength Prediction and Defect Detection in Data-Constrained Training Environments.” Sensors (Basel) 24 (20): 6553. https://doi.org/10.3390/s24206553.10.3390/s24206553 ↗
16. 16. Wang, X., and X. Yu. 2023. “Understanding the effect of transfer learning on the automatic welding defect detection.” NDT & E International 134: 102784. https://doi.org/10.1016/j.ndteint.2022.102784.10.1016/j.ndteint.2022.102784 ↗
17. 17. Lu, R., Y. Zhang, Q. Huang, P. Zeng, Z. Shi, and S. Ye. 2024. “A refined edge-aware convolutional neural networks for agricultural parcel delineation.” International Journal of Applied Earth Observation and Geoinformation 133: 104084. https://doi.org/10.1016/j.jag.2024.104084.10.1016/j.jag.2024.104084 ↗
18. 18. Liu, F., Z. Hua, J. Li, and L. Fan. 2023. “MFBGR: Multi-scale feature boundary graph reasoning network for polyp segmentation.” Engineering Applications of Artificial Intelligence 123: 106213. https://doi.org/10.1016/j.engappai.2023.106213.10.1016/j.engappai.2023.106213 ↗
19. 19. Zou, N., Z. Xiang, Y. Chen, S. Chen, and C. Qiao. 2019. “Boundary-aware CNN for semantic segmentation.” IEEE Access 7: 114520–114528. https://doi.org/10.1109/ACCESS.2019.2935816.10.1109/access.2019.2935816 ↗
20. 20. Chen, Z., T. Xu, Y. Pan, N. Shen, H. Chen, and J. Li. 2024. “Edge feature enhancement for fine-grained segmentation of remote sensing images.” IEEE Transactions on Geoscience and Remote Sensing 62: 1–13. https://doi.org/10.1109/TGRS.2024.3443247.10.1109/tgrs.2024.3443247 ↗
21. 21. Ferguson, M., R. Ak, Y. T. T. Lee, and K. H. Law. 2018. “Detection and segmentation of manufacturing defects with convolutional neural networks and transfer learning.” Smart and Sustainable Manufacturing Systems 2 (1): 137–64. https://doi.org/10.1520/SSMS20180033.10.1520/ssms20180033 ↗
22. 22. Ajmi, C., J. Zapata, J. J. Martínez-Álvarez, G. Doménech, and R. Ruiz. 2020. “Using deep learning for defect classification on a small weld X-ray image dataset.” Journal of Nondestructive Evaluation 39 (68): 1–13. https://doi.org/10.1007/s10921-020-00719-9.10.1007/s10921-020-00719-9 ↗
23. 23. Zhang, L., H. Pan, B. Jia, L. Li, M. Pan, and L. Chen. 2025. “Lightweight DCGAN and MobileNet-based model for detecting X-ray welding defects under unbalanced samples.” Scientific Reports 15 (1): 6221. https://doi.org/10.1038/s41598-025-89558-0.10.1038/s41598-025-89558-0 ↗
24. 24. Tang, D., J. Zhang, P. Wang, and Y. He. 2025. “Weld TOFD defect classification method based on multi-scale CNN and cascaded focused attention.” Journal of Manufacturing Processes 138:157–68. https://doi.org/10.1016/j.jmapro.2025.02.019.10.1016/j.jmapro.2025.02.019 ↗
25. 25. Li, L., P. Wang, Y. Li, Z. Lü, Y. Xu, R. Di, X. Li, and T. Geng. 2025. “High-resolution weld defect detection with RSU-MLP and dynamic kernel supervision.” Measurement: Journal of the International Measurement Confederation 242: 116208. https://doi.org/10.1016/j.measurement.2024.116208.10.1016/j.measurement.2024.116208 ↗
26. 26. Li, Y. Q., S. Z. Li, G. L. Sun, and Y. Pu. 2025. “Lightweight Swin Transformer combined with multi-scale feature fusion for face expression recognition.” Opto-Electronic Engineering 52 (1): 240234. https://doi.org/10.12086/oee.2025.240234 [in Chinese].
27. 27. Ju, R. Y., C. C. Chen, J. S. Chiang, Y. S. Lin, and W. H. Chen. 2023. “Resolution enhancement processing on low-quality images using Swin Transformer based on interval dense connection strategy.” Multimedia Tools and Applications 83 (5): 14839–14855. https://doi.org/10.1007/s11042-023-16088-0.10.1007/s11042-023-16088-0 ↗
28. 28. Zhu, X., X. Huang, W. Cao, X. Yang, Y. Zhou, and S. Wang. 2024. “Road Extraction from Remote Sensing Imagery with Spatial Attention Based on Swin Transformer.” Remote Sensing (Basel) 16 (7): 1183. https://doi.org/10.3390/rs16071183.10.3390/rs16071183 ↗