Automatic segmentation of microporous defects in composite film materials based on the improved attention U-Net module
Source abstract
Abstract
Composite film materials offer several advantages, such as high strength-to-weight ratios, flexibility, durability, or cost-effectiveness, which make them ideal for a wide range of applications across various industries. However, microporous defects in the material compromise its durability and air-tightness. Thus far, few works have been performed on inquiring non-destructive testing methods intended for the detection of microporous defects in composite film materials. In this study, the automatic detection of microporous defects is achieved by using a thermal imaging inspection system developed on the basis of a laser infrared system and the U-Net. Innovatively, an improved attention module is added to U-Net to boost its detection performance. While using a laser as a heat source, a fibre laser beam shaper is chosen as the laser beam expansion to ensure a uniform intensity distribution of the laser beam within a small heating area. Experimental results show that the proposed method can reliably detect microporous defects in the composite film materials with sizes in the order of millimeters/micrometers.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Thermography and Photoacoustic Techniques · Industrial Vision Systems and Defect Detection · Photoacoustic and Ultrasonic Imaging
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Harbin Institute of Technology
Central South University
Tomsk Polytechnic University
Funding
National Key Laboratory of Science and Technology on Advanced Composites in Special Environments · Russian Scientific Foundation
References (35)
Show source-provided references
1. 10.1631/jzus.A110033610.1631/jzus.A1100336 ↗
2. 10.2514/1.C03668410.2514/1.C036684 ↗
3. 10.1109/AUCC.2016.786820910.1109/AUCC.2016.7868209 ↗
4. 10.1109/ACC.2016.752660310.1109/ACC.2016.7526603 ↗
5. 10.1109/TAC.2015.250356610.1109/TAC.2015.2503566 ↗
6. 10.1109/TIE.2014.230177310.1109/TIE.2014.2301773 ↗
7. 10.1109/CCA.2014.698146410.1109/CCA.2014.6981464 ↗
8. 10.1109/ICIEA.2010.551548110.1109/ICIEA.2010.5515481 ↗
9. 10.1016/j.culher.2023.06.01810.1016/j.culher.2023.06.018 ↗
10. 10.1016/j.ndteint.2018.11.01010.1016/j.ndteint.2018.11.010 ↗
11. 10.1080/17686733.2022.215225910.1080/17686733.2022.2152259 ↗
12. 10.1080/17686733.2021.202501510.1080/17686733.2021.2025015 ↗
13. 10.1016/j.ndteint.2016.11.01210.1016/j.ndteint.2016.11.012 ↗
14. 10.1016/j.ndteint.2013.07.00210.1016/j.ndteint.2013.07.002 ↗
15. 10.1109/TII.2015.247985610.1109/TII.2015.2479856 ↗
16. 10.1007/s10921-022-00845-610.1007/s10921-022-00845-6 ↗
17. 10.1016/j.infrared.2020.10328210.1016/j.infrared.2020.103282 ↗
18. 10.1016/j.ndteint.2021.10255010.1016/j.ndteint.2021.102550 ↗
19. 10.1080/17686733.2022.203352910.1080/17686733.2022.2033529 ↗
20. 10.1080/17686733.2022.203353110.1080/17686733.2022.2033531 ↗
21. 10.1080/17686733.2022.215867810.1080/17686733.2022.2158678 ↗
22. 10.3390/s2102039510.3390/s21020395 ↗
23. 10.1016/j.ndteint.2019.10216410.1016/j.ndteint.2019.102164 ↗
24. 10.1016/j.compscitech.2022.10952610.1016/j.compscitech.2022.109526 ↗
25. Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference Proceedings, Part III; Springer International Publishing; 2015. p. 234–241. Munich, Germany.
26. 10.1038/nature1453910.1038/nature14539 ↗
27. 10.1016/j.ndteint.2019.10214710.1016/j.ndteint.2019.102147 ↗
28. 10.1080/17686733.2022.212913510.1080/17686733.2022.2129135 ↗
29. 10.1109/ICCV.2015.16910.1109/ICCV.2015.169 ↗
30. 10.1109/CVPR.2018.0074510.1109/CVPR.2018.00745 ↗
31. 10.1109/CVPR42600.2020.0115510.1109/CVPR42600.2020.01155 ↗
32. 10.1109/CVPR46437.2021.0135010.1109/CVPR46437.2021.01350 ↗
33. 10.1109/ICCV.2019.0092610.1109/ICCV.2019.00926 ↗
34. 10.1007/978-3-030-01234-2_110.1007/978-3-030-01234-2_1 ↗
35. IR View Research Edition. Laval University Version [software]. 2018. Available from: http://www.visiooimage.com-office@visiooimage.com