HyDD: Hybrid Defect Detection Framework for Inspection Using X-ray CT Data
Keywords:
image processing, computerized tomography (CT), nondestructive testing, industrial inspection, classification, deep learning, annotationAbstract
In most manufacturing and assembly industries, such as automotive, aerospace, and nuclear power plants, it is critical to inspect the integrity of the used parts and components to avoid possible system failure for safety and quality control reasons. In addition, an unexpected failure causes extra costs due to the additional costs of maintenance, reparations, or even service delivery delays. Therefore, building an automated inspection system using industrial testing techniques such as Non-Destructive Testing (NDT) using X-ray computerized tomography (CT) is highly recommended. This paper proposes a Hybrid Defect Detection (HyDD) to automate the inspection process and improve defect detection accuracy. The proposed framework investigates incorporating deep learning and image processing techniques to enhance inspection performance and build an adaptive inspection system. The proposed method is validated using a real X-ray CT dataset with different types of defect scenarios. The results show that the proposed method could detect most defects and achieve up to 88% of the average detection rate. The proposed framework could be extended to defect-type recognition using object detection or semantic segmentation deep learning models.
References
[1] B.-J. MacKinnon, “Point Lepreau nuclear plant taken offline after power loss | CBC News.” Dec. 2022. Available: https://www.cbc.ca/news/canada/new-brunswick/point-lepreau-offline-power-loss-canadian-nuclear-safety-commission-nb-power-1.6686212
[2] D. Chauveau, “Review of NDT and process monitoring techniques usable to produce high-quality parts by welding or additive manufacturing,” Welding in the World, vol. 62, pp. 1097–1118, Sep. 2018, doi: 10.1007/S40194-018-0609-3/FIGURES/37.
[3] R. Halmshaw, “Non-destructive testing: Metallurgy and materials science series,” London: Edward Arnold, 1987.
[4] H. Masoom, R. S. Adve, and R. S. Cobbold, “Target detection in diagnostic ultrasound: Evaluation of a method based on the CLEAN algorithm,” Ultrasonics, vol. 53, no. 2, pp. 335–344, 2013.
[5] S. Akcay and T. Breckon, “Towards automatic threat detection: A survey of advances of deep learning within X-ray security imaging,” Pattern Recognition, vol. 122, p. 108245, 2022, doi: 10.1016/j.patcog.2021.108245.
[6] D. Mery and C. Arteta, “Automatic defect recognition in x-ray testing using computer vision,” in 2017 IEEE winter conference on applications of computer vision (WACV), 2017, pp. 1026–1035. doi: 10.1109/WACV.2017.119.
[7] D.-M. Tsai and P.-H. Jen, “Autoencoder-based anomaly detection for surface defect inspection,” Advanced Engineering Informatics, vol. 48, p. 101272, Apr. 2021, doi: 10.1016/j.aei.2021.101272.
[8] I. Szabo, J. Sun, G. Feng, J. Kanfoud, T. H. Gan, and C. Selcuk, “Automated defect recognition as a critical element of a three dimensional x-ray computed tomography imaging-based smart non-destructive testing technique in additive manufacturing of near net-shape parts,” Applied Sciences 2017, Vol. 7, Page 1156, vol. 7, p. 1156, Nov. 2017, doi: 10.3390/APP7111156.
[9] X. Zhao and D. W. Rosen, “Simulation study on evolutionary cycle to cycle time control of exposure controlled projection lithography,” Rapid Prototyping Journal, 2016.
[10] G. Dong, J. Marleau-Finley, and Y. F. Zhao, “Investigation of electrochemical post-processing procedure for ti-6Al-4V lattice structure manufactured by direct metal laser sintering (DMLS),” The International Journal of Advanced Manufacturing Technology, vol. 104, no. 9, pp. 3401–3417, 2019.
[11] “Speed|scan CT 64 | High-Speed CT Scanner.” https://www.bakerhughesds.com/industrial-x-ray-ct-scanners/speedscan-ct-64.
[12] “Series Inspection CT | X-ray CT systems | Nikon Metrology.” https://www.nikonmetrology.com/en-us/x-ray-ct/x-ray-ct-systems-series-inspection-ct.
[13] “YXLON - YXLON X-ray and CT inspection systems.” https://www.yxlon.com/en/products/x-ray-and-ct-inspection-systems/yxlon-ux20.
[14] N. Hashem, M. Pryor, D. Haas, and J. Hunter, “Design of a computed tomography automation architecture,” vol. 11. 2021. doi: 10.3390/app11062858.
[15] “XRHRobotStar - In-line ADR X-ray inspection.” https://visiconsult.de/products/ndt/xrhrobotstar/.
[16] N. Brierley, “Mapping performance of CT,” International Symposium on Digital Industrial Radiology and Computed Tomography, 2019.
[17] L. A. Feldkamp, L. C. Davis, and J. W. Kress, “Practical cone-beam algorithm,” Journal of the Optical Society of America A, vol. 1, no. 6, p. 612, Jun. 1984, doi: 10.1364/josaa.1.000612.
[18] Y. Lei et al., “4D-CT deformable image registration using multiscale unsupervised deep learning,” Physics in Medicine and Biology, vol. 65, p. 085003, Apr. 2020, doi: 10.1088/1361-6560/ab79c4.
[19] H. A. Gabbar, A. Chahid, M. J. A. Khan, O. G. Adegboro, and M. I. Samson, “CTIMS: Automated Defect Detection Framework Using Computed Tomography,” Applied Sciences (Switzerland), vol. 12, no. 4, p. 2175, 2022, doi: 10.3390/app12042175.
[20] R. Kuprieiev et al., “DVC: Data Version Control - Git for Data & Models,” 2023, doi: 10.5281/ZENODO.11445998.
[21] A.L Clarke et al., “Development of an automated digital radiography system for the non-destructive inspection of green powder metallurgy parts” Euro PM 2014 International Conference and Exhibition (2014). European Powder Metallurgy Association (EPMA)
[22] M., Krumm et al., “Rapid Robotic X - ray Computed Tomography of Large Assemblies in Automotive Production”. 8th Conference on Industrial Computed Tomography, Wels, Austria (iCT 2018), pp. 1–9 (2018). www.3dct.at
Downloads
Published
Issue
Section
License
Copyright (c) 2025 International Journal of Natural Sciences: Current and Future Research Trends
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who submit papers with this journal agree to the following terms.
