A Variable Attention Nested UNet++ Network-Based NDT X-ray Image Defect Segmentation Method
Source abstract
Abstract
In this paper, we describe a new method for non-destructive testing (NDT) X-ray image defect segmentation by introducing a variable attention nested UNet++ network. To further enhance the performance of the faint defect extraction and its clear visibility, a pre-processing method based on pyramid model is also added to the proposed method to effectively perform high dynamic range compression and defect enhancement on the 16-bit raw image. To illustrate its effectiveness and efficiency, we applied the proposed algorithm to the X-ray image defect segmentation problem and carried out extensive experiments. The results support that the proposed method outperforms the existing representative techniques in extracting defect for real X-ray images collected directly from industrial lines, which achieves the better performance with 89.24% IoU, and 94.31% Dice.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Integrated Circuits and Semiconductor Failure Analysis · Advanced X-ray and CT Imaging · Advanced Neural Network Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Image and A.I. Laboratory, Department of Electronics Engineering, Pusan National University, Busan 46241, Korea
Funding
Guangdong Science and Technology Project · Creative Human Resource Education and Research Programs for ICT Convergence
References (46)
Show source-provided references
1. Automatic Defect Detection in Aluminium Castings using Radioscopic Image Sequences
2. Nondestructive observation of pores during press forming of aluminum foam by X-ray radiography10.1016/j.matchar.2020.110631 ↗
3. Non Destructive Testing For Detection abnormal Object in The X-Ray Images10.1088/1742-6596/1660/1/012104 ↗
4. Automated flaw detection in aluminum castings based on the tracking of potential defects in a radioscopic image sequence10.1109/TRA.2002.805646 ↗
5. Automated radioscopic testing of aluminum die castings
6. State-of-the-art of weld seam inspection by radiographic testing
7. X-ray systems for security and industrial inspection10.1108/02602280810882535 ↗
8. Flaw detection in aluminium die castings using simultaneous combination of multiple views
9. Automated fish bone detection using X-ray imaging10.1016/j.jfoodeng.2011.03.007 ↗
10. Inspection of complex objects using multiple-X-ray views10.1109/TMECH.2014.2311032 ↗
11. Automatic localization and segmentation of the ventricles in magnetic resonance images10.1109/TCSVT.2020.2981530 ↗
12. A new approach for segmentation and quantification of cells or nanoparticles10.1109/TII.2016.2542043 ↗
13. Simulation of defects in aluminium castings using CAD models of flaws and real X-ray images
14. Mery, D., and Arteta, C. (2017, January 24–31). Automatic defect recognition in x-ray testing using computer vision. Proceedings of the 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, USA.10.1109/WACV.2017.119 ↗
15. Mery, D., and Pieringer, C. (2021). Computer Vision for X-ray Testing: Imaging, Systems, Image Databases, and Algorithms, Springer International Publishing. [2nd ed].10.1007/978-3-030-56769-9 ↗
16. Deep learning for generic object detection: A survey10.1007/s11263-019-01247-4 ↗
17. Mery, D., and Pieringer, C. (2021). Deep Learning in X-ray Testing. Computer Vision for X-ray Testing, Springer.10.1007/978-3-030-56769-9 ↗
18. Aluminum casting inspection using deep object detection methods and simulated ellipsoidal Defects10.1007/s00138-021-01195-5 ↗
19. Tokime, R., Maldague, X., and Perron, L. (2019, January 18–20). Automatic Defect Detection for X-Ray inspection: Identifying defects with deep convolutional network. Proceedings of the Canadian Institute for Non-destructive Evaluation (CINDE), Edmonton, AB, Canada.
20. Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for Biomedical Image Segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer.10.1007/978-3-319-24574-4_28 ↗
21. H-DenseUNet: Hybrid densely connected UNet for liver and tumor segmentation from CT volumes10.1109/TMI.2018.2845918 ↗
22. Tokime, R.B., Maldague, X., and Perron, L. (2019, January 23–25). Automatic Defect Detection for X-ray inspection: A U-Net approach for defect segmentation. Proceedings of the Digital Imaging and Ultrasonics for NDT Conference 2019, New Orleans, LA, USA.
23. Zhou, Z., Rahman Siddiquee, M.M., Tajbakhsh, N., and Liang, J. (2018). Unet++: A nested u-net architecture for medical image segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, Springer.10.1007/978-3-030-00889-5_1 ↗
24. Unet++: Redesigning skip connections to exploit multiscale features in image segmentation10.1109/TMI.2019.2959609 ↗
25. Li, C., Tan, Y., Chen, W., Luo, X., Gao, Y., Jia, X., and Wang, Z. (2020, January 25–28). Attention unet++: A nested attention-aware u-net for liver ct image segmentation. Proceedings of the 2020 IEEE International Conference on Image Processing (ICIP), Abu Dhabi, United Arab Emirates.10.1109/ICIP40778.2020.9190761 ↗
26. Gite, S., Mishra, A., and Kotecha, K. (2022). Enhanced lung image segmentation using deep learning. Neural Comput. Appl., 1–15.10.1007/s00521-021-06719-8 ↗
27. LAEDNet: A Lightweight Attention Encoder–Decoder Network for ultrasound medical image segmentation10.1016/j.compeleceng.2022.107777 ↗
28. COVID-19 infection localization and severity grading from chest X-ray images10.1016/j.compbiomed.2021.105002 ↗
29. Diniz, J.O.B., Dias Júnior, D.A., da Cruz, L.B., da Silva, G.L.F., and Ferreira, J.L. (Comput. Methods Biomech. Biomed. Eng. Imaging Vis., 2022). Heart segmentation in planning CT using 2.5 D U-Net++ with attention gate, Comput. Methods Biomech. Biomed. Eng. Imaging Vis., in press.10.1080/21681163.2022.2043779 ↗
30. Moon Impact Crater Detection Using Nested Attention Mechanism Based UNet++10.1109/ACCESS.2021.3066445 ↗
31. Yang, X., Li, X., Ye, Y., Zhang, X., Zhang, H., Huang, X., and Zhang, B. (2019, January 14–19). Road detection via deep residual dense u-net. Proceedings of the 2019 International Joint Conference on Neural Networks (IJCNN), Budapest, Hungary.10.1109/IJCNN.2019.8851728 ↗
32. Liu, B., Zhang, X., Gao, Z., and Chen, L. (2017). Weld defect images classification with vgg16-based neural network. International Forum on Digital TV and Wireless Multimedia Communications, Springer.10.1007/978-981-10-8108-8_20 ↗
33. Nondestructive defect detection in castings by using spatial attention bilinear convolutional neural network10.1109/TII.2020.2985159 ↗
34. Lin, T., Dollár, P., Girshick, R., and He, K. (2017, January 21–26). Feature pyramid networks for object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.106 ↗
35. Approaches for improvement of the X-ray image defect detection of automobile casting aluminum parts based on deep learning10.1016/j.ndteint.2019.102144 ↗
36. Automated detection of defects with low semantic information in X-ray images based on deep learning10.1007/s10845-020-01566-1 ↗
37. Spatial pyramid pooling in deep convolutional networks for visual recognition10.1109/TPAMI.2015.2389824 ↗
38. He, K., Gkioxari, G., Dollar, P., and Girshick, R. (2017, January 22–29). Mask R-CNN. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.10.1109/ICCV.2017.322 ↗
39. Data augmentation for X-ray prohibited item images using generative adversarial networks10.1109/ACCESS.2019.2902121 ↗
40. Aluminum casting inspection using deep learning: A method based on convolutional neural networks10.1007/s10921-020-0655-9 ↗
41. Defect image sample generation with GAN for improving defect recognition
42. GDXray: The database of X-ray images for nondestructive testing10.1007/s10921-015-0315-7 ↗
43. (2022, April 02). Deepsea Precision Co., Ltd. Shenzhen City, China. Available online: http://www.shenhaijingmi.com/.
44. (2022, April 01). ASTM E2339-15. Standard Practice for Digital Imaging and Communication in Nondestructive Evaluation (DICONDE), 2015 Edition, Available online: https://www.astm.org/e2339-15.html.
45. Intel Corporation (2022, April 01). Stylized as Intel, Is an American Multinational Corporation and Technology Company Headquartered in Santa Clara, California, USA. Available online: https://www.intel.com/.
46. Nvidia (2022, April 01). Nvidia Corporation Is an American multinational Technology Company Incorporated in Delaware and Based in Santa Clara, California, USA. Available online: https://www.nvidia.com/.