Abstract
Non-destructive periodic monitoring is crucial to evaluate the state of welded joints in many different types of structures over time. One way to do this is to compare new ultrasound scans of a weld with previous ones, which allows qualified analysts to verify any signs of degradation. Considering the very large number of scans to be analyzed and the rather low number of certified analysts, investigating methods to automate this task becomes very relevant. This study proposes a novel approach that uses a 3D geometric model of welds and their defects, generated directly from the raw ultrasound data, to create a unique signature of each weld in a database, allowing retrieval of previous scans when the same weld is re-scanned. Three experiments were designed to test the recognition and retrieval efficiency using this process. The first experiment is a case study that tests if the algorithm can retrieve a list of similar scans. The second experiment defines the threshold value used to determine whether two scans belong to the same weld. The last experiment is designed to find the configuration of the recognition algorithm that yields the best performance. Validation tests using a database of real industrial 3D scans show that the proposed approach can efficiently recognize different scans of the same weld.
Similar content being viewed by others
References
American Society for Nondestructive Testing (2007) Nondestructive testing handbook, third edition: volume 7, ultrasonic testing (UT). American Society for Nondestructive Testing, Inc., United States of America, p 588
Taheri H et al (2019) Investigation of nondestructive testing methods for friction stir welding. Metals 9(6):624
Luo H, Chen QH, Doan NCN, Lin W (2016) Automated identification and characterization of clustered weld defects. In: 2016 IEEE International Conference on Advanced Intelligent Mechatronics (AIM). IEEE, pp 536–541
Groover MP (2007) Fundamentals of modern manufacturing: materials processes, and systems. John Wiley & Sons
Oh SB, Cheong YM, Lee DH, Kim KM (2017) Development of a magnetostrictive guided wave technique for defect detection and monitoring in a pipe weld. J Korean Soc Nondestruct Test 37(4):230–238
van Sloun RJ, Cohen R, Eldar YC (2019) Deep learning in ultrasound imaging. Proceedings of the IEEE 108(1):11–29
Munir N, Park J, Kim H-J, Song S-J, Kang S-S (2020) Performance enhancement of convolutional neural network for ultrasonic flaw classification by adopting autoencoder. NDT E Int 101:102218
Cantero-Chinchilla S, Wilcox PD, Croxford AJ (2022) Deep learning in automated ultrasonic NDE-developments, axioms and opportunities. NDT & E International 102703
Zhang Z et al (2022) Deep learning-based monitoring of surface residual stress and efficient sensing of AE for laser shock peening. J Mater Process Technol 303:117515
Smoqi Z et al (2022) Monitoring and prediction of porosity in laser powder bed fusion using physics-informed meltpool signatures and machine learning. J Mater Process Technol 304:117550
Ness KL, Paul A, Sun L, Zhang Z (2022) Towards a generic physics-based machine learning model for geometry invariant thermal history prediction in additive manufacturing. J Mater Process Technol 302:117472
Maier M, Kunstmann H, Zwicker R, Rupenyan A, Wegener K (2022) Autonomous and data-efficient optimization of turning processes using expert knowledge and transfer learning. J Mater Process Technol 303:117540
Molitor DA, Kubik C, Becker M, Hetfleisch RH, Lyu F, Groche P (2022) Towards high-performance deep learning models in tool wear classification with generative adversarial networks. J Mater Process Technol 302:117484
Yang J, Lu X, Chen W (2021) A robust scheme for 3D point cloud copy detection. arXiv preprint arXiv 2110.00972
Shukla D, Sharma M (2012) Watermarking schemes for copy protection: a survey. Int J Comput Sci Eng Surv 3(1):65
Medimegh N, Belaid S, Werghi N (2015) A survey of the 3D triangular mesh watermarking techniques. Int J Multimed 1(1)
Cheng SW, Dey TK., Shewchuk J (2012) Delaunay Mesh Generation. CRC Press
Pang L, Lan Y, Guo J, Xu J, Xu J, Cheng X (2017) Deeprank: a new deep architecture for relevance ranking in information retrieval. Association for Computing Machinery, New York, NY. https://doi.org/10.1145/3132847.3132914
Wang J et al (2014) Learning fine-grained image similarity with deep ranking. Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, USA, pp 1386–1393
Sinha P, Russell R (2011) A perceptually based comparison of image similarity metrics. Perception 40(11):1269–1281
Morra L, Lamberti F (2019) Benchmarking unsupervised near-duplicate image detection. Expert Syst Appl 135:313–326
Naumann F, Herschel M (2010) An introduction to duplicate detection. Synth Lect Data Manag 2:1–87
Liu X, Xu L (2013) Detecting approximately duplicate records in database. In: Proceedings of International Conference on Engineering, Science and Applications (IEA) 2012. Springer, pp 325–332
Xu H, Wang J, Li Z, Zeng G, Li S, Yu N (2011) Complementary hashing for approximate nearest neighbor search. In: 2011 Int Conf Comput Vision, IEEE, pp 1631–1638
Paulevé L, Jégou H, Amsaleg L (2010) Locality sensitive hashing: a comparison of hash function types and querying mechanisms. Pattern Recogn Lett 31:1348–1358
Bentley JL (1975) |Multidimensional binary search trees used for associative searching. Commun ACM 18(9):509–517
Ciaccia P, Patella M, Zezula P (1997) M-tree: an efficient access method for similarity search in metric spaces. In Vldb Citeseer 97:426–435
Malkov YA, Yashunin DA (2018) Efficient and robust approximate nearest neighbor search using hierarchical navigable small world graphs. IEEE Trans Pattern Anal Mach Intell 42(4):824–836
Aumüller M, Bernhardsson E, Faithfull A (2017) ANN-benchmarks: a benchmarking tool for approximate nearest neighbor algorithms. In: International Conference on Similarity Search and Applications. Springer, pp 34–49
Ke Y, Sukthankar R, Huston L, Ke Y, Sukthankar R (2004) Efficient near-duplicate detection and sub-image retrieval. ACM Int Conf Multimed 4(1):5
Provencal E, Laperrière L (2021) Detection of exact and near duplicates in phased-array ultrasound weld scan. Procedia Manuf 54:263–268
Provencal E, Laperrière L (2022) WeldNet: from 3D phased-array ultrasound scans to 3D geometrical models of welds and defects. CIRP Ann 71(1):445–448
Erler P, Guerrero P, Ohrhallinger S, Mitra NJ, Wimmer M (2020) Points2surf learning implicit surfaces from point clouds. European Conference on Computer Vision. Springer, pp 108–124
Aspert N, Santa-Cruz D, Ebrahimi T (2002) Mesh: measuring errors between surfaces using the hausdorff distance. In Proc IEEE Int Conf Multimed Expo IEEE 1:705–708
Ma J et al (2021) Loss odyssey in medical image segmentation. Med Image Anal 71:102035
Provencal E, Laperrière L (2021) Identification of weld geometry from ultrasound scan data using deep learning. Procedia CIRP 104:122–127
Funding
This work is supported by the Natural Sciences and Engineering Council of Canada (NSERC), [401221093] and the Mitacs accelerate program [IT15174].
Author information
Authors and Affiliations
Contributions
All authors contributed to the study’s conception, design, and analysis. The first draft of the manuscript was written by Etienne Provencal, and then Luc Laperrière commented on and revised the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Provencal, E., Laperrière, L. Periodical monitoring of 3D welds and defects generated from ultrasound scans. Int J Adv Manuf Technol 125, 1239–1249 (2023). https://doi.org/10.1007/s00170-022-10785-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00170-022-10785-0