Various nondestructive testing (NDT) methods have been proposed to detect defects inside products. The hammering test is an NDT technique widely used for this purpose. In this test method, a worker judges whether a part is defective or not by listening to the sound after hitting the product with a hammer. Conventional research has shown that a classifier using machine learning can discriminate the hammering data with high accuracy. However, to use these machine learning methods, a lot of samples are needed for learning. In actual industrial situations, it is difficult to collect a lot of samples of defective products. Regarding the hammering test, a machine learning method that can correctly discriminate defective products without sample data has not been proposed. This study aims to construct a system that can correctly discriminate the hammering test data even when there are no defective samples. We propose a method using ‘transfer learning.’ We conducted case studies to demonstrate the effectiveness of the proposed method using two variants of a brazed product. First, we verified the effectiveness of normal machine learning in a hammering test. In this study, we succeeded in discriminating brazed products, which were not correctly discriminated by the workers. We then applied the proposed method to brazed products. We succeeded in discriminating a variant of the brazed products by transferring the knowledge learned from another variant of the brazed products.

1. [1] K. Cheng and B. J. Richard, “e-Manufacturing: Characteristics, applications and potentials,” Progress in Natural Science, Vol.18, pp. 1323-1328, 2008.
2. [2] S. Ravikumar, K. I. Ramachandran, and V. Sugumaran, “Machine learning approach for automated visual inspection of machine components,” Expert Systems with Applications, Vol.38, pp. 3260-3266, 2011.
3. [3] V. Azamfirei, A. Granlund, and Y. Lagrosen, “Multi-Layer Quality Inspection System Framework for Industry 4.0,” Int. J. Automation Technol., Vol.15, No.5, pp. 641-650, 2021.
4. [4] K. Liu et al., “Steel surface defect detection using a new Haar-Weibull-variance model in unsupervised manner,” IEEE Trans. on Instrumentation and Measurement, Vol.6610, pp. 2585-2596, 2017.
5. [5] Z. Zhao et al., “A Surface Defect Detection Method Based on Positive Samples,” Proc. of the Pacific Rim Int. Conf. on Artificial Intelligence, pp. 473-481, 2018.
6. [6] X. Tao et al., “Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks,” Applied Sci., Vol.8, pp. 1575-1589, 2018.
7. [7] Z. Xiao et al., “Defect detection and classification of galvanized stamping parts based on fully convolution neural network,” Proc. SPIE 10615, 9th Int. Conf. on Graphic and Image Processing (ICGIP 2017), 106150K, 2018.
8. [8] S. Y. Jung et al., “Defect Detection on Randomly Textured Surfaces by Convolutional Neural Networks,” Proc. of the 2018 IEEE/ASME Int. Conf. on Advanced Intelligent Mechatronics (AIM), pp. 1456-1461, 2018.
9. [9] C. Li and X. Chen, “Surface Defect Detection Algorithm Based on Local Neighborhood Analysis,” V. E. Balas, L. C. Jain, and X. Zhao (Eds.), “Information Technology and Intelligent Transportation Systems,” pp. 307-315, Springer, 2017.
10. [10] M. Zhang et al., “The application of one-class classifier based on CNN in image defect detection,” Procedia Computer Sci. Vol.114, pp. 341-348, 2017.
11. [11] S. Mei, H. Yang, and Z. Yin, “An Unsupervised-Learning-Based Approach for Automated Defect Inspection on Textured Surfaces,” IEEE Trans. on Instrumentation and Measurement, Vol.67, No.6, pp. 1266-1277, 2018.
12. [12] D. Weimer, B. Scholz-Reiter, and M. Shpitalni, “Design of deep convolutional neural network architectures for automated feature extraction in industrial inspection,” CIRP Ann., Vol.65, pp. 417-420, 2016.
13. [13] https://www.un.org/sustainabledevelopment/ [Accessed May 9, 2022]
14. [14] European Commission, “Industry 5.0: A Transformative Vision for Europe,” ESIR Policy Brief No.3, European Union, 2022.
15. [15] B. Hadorn, M. Courant et al., “Towards Human-Centered Cyber-Physical Systems: A Modeling Approach,” The University of Fribourg, 2016.
16. [16] F. Ansari, M. Khobreh et al., “A problem-solving ontology for human-centered cyber physical production systems,” CIRP J. of Manuf. Sci. and Tech., Vol.22, pp. 91-106, 2018.
17. [17] S. Ghorai et al., “Automatic defect detection on hot-rolled flat steel products,” IEEE Trans. on Instrumentation and Measurement, Vol.62, pp. 612-621, 2013.
18. [18] X. Li et al., “Improving automatic detection of defects in castings by applying wavelet technique,” IEEE Trans. on Industrial Electronics, Vol.53, pp. 1927-1934, 2006.
19. [19] Y. V. Chugui et al., “Optical Measuring and Laser Technologies for Scientific and Industrial Applications,” Int. J. Automation Technol., Vol.9, No.5, pp. 515-524, 2015.
20. [20] T. R. Naik, M. Malhotra, and J. S. Popovics, “The ultrasonic pulse velocity method,” V. M. Malhotra and N. J. Carino (Eds.), “Handbook on Nondestructive Testing of Concrete,” pp. 182-200, CRC Press, 2003.
21. [21] M. J. Sansalone and W. B. Streett, “Impact-echo. Nondestructive evaluation of concrete and masonry,” Bullbrier Press, 1997.
22. [22] N. Blindow et al., “Ground penetrating radar,” K. Knödel, G. Lange, and H.-J. Voigt, “Environmental Geology,” pp. 283-335, Springer, 2007.
23. [23] I. Z. Abidin et al., “Advantages and applications of eddy current thermography testing for comprehensive and reliable defect assessment,” Proc. of 18th World Conf. on Nondestructive Testing, Vol.1620, 2012.
24. [24] T. Azuma, R. Ito, S. Soma, S. Murakami, and T. Kuriyagawa, “Development of Non-Destructive Technology for Detecting Grinding Burns,” Int. J. Automation Technol., Vol.7, No.6, pp. 700-707, 2013.
25. [25] Y. Umehara and N. Moronuki, “Nondestructive Inline Inspection of Through-Silicon Vias Based on X-Ray Imaging and its Uncertainty Budget,” Int. J. Automation Technol., Vol.13, No.2, pp. 289-300, 2019.
26. [26] A. Nair and C. S. Cai, “Acoustic emission monitoring of bridges: Review and case studies,” Eng. Structures, Vol.32, pp. 1704-1714, 2010.
27. [27] H. Fujii, A. Yamashita, and H. Asama, “Improvement of environmental adaptivity of defect detector for hammering test using boosting algorithm,” Proc. of the 2015 IEEE/RSJ Int. Conf. on Intelligence Robots and Systems (IROS), pp. 6507-6514, 2015.
28. [28] C. Gomez-Cardenas et al., “New optimization algorithm for optimal spatial sampling during non-destructive testing of concrete structures,” Eng. Structures, Vol.88, pp. 92-99, 2015.
29. [29] M. Kemmler et al., “One-class classification with Gaussian processes,” Pattern Recognition, Vol.46, pp. 3507-3518, 2013.
30. [30] B. Krawczyk, M. Woźnia, and F. Herrera, “Weighted one-class classification for different types of minority class examples in imbalanced data,” Proc. of the 2014 IEEE Symp. on Computational Intelligence and Data Mining (CIDM), pp. 337-344, 2014.
31. [31] M. A. Mazurowski et al., “Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance,” Neural Networks, Vol.21, pp. 427-436, 2008.
32. [32] V. Ganganwar, “An overview of classification algorithms for imbalanced datasets,” Int. J. of Emerging Tech. and Advanced Eng., Vol.2, pp. 42-47, 2012.
33. [33] Q. Leng et al., “One-class classification with extreme learning machine,” Mathematical Problems in Eng., Vol.2015, 412957, 2015.
34. [34] Y. Sun et al., “Cost-sensitive boosting for classification of imbalanced data,” Pattern Recognition, Vol.40, pp. 3358-3378, 2007.
35. [35] D. Wang, D. S. Yeung, and E. C. C. Tsang, “Structured one-class classification,” IEEE Trans. on Systems, Man, and Cybernetics, Part B (Cybernetics), Vol.36, pp. 1283-1295, 2006.
36. [36] Y. Sun, A. K. C. Wong, and M. S. Kamel, “Classification of imbalanced data: A review,” Int. J. of Pattern Recognition and Artificial Intelligence, Vol.23, pp. 687-719, 2009.
37. [37] C. Huang et al., “Learning deep representation for imbalanced classification,” Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, pp. 5375-5384, 2016.
38. [38] V. López et al., “An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics,” Inf. Sci., Vol.250, pp. 113-141, 2013.
39. [39] P. Juszczak, “Learning to recognise: A study on one-class classification and active learning,” Delft University of Technology, 2006.
40. [40] S. J. Pan and Q. Yang, “A survey on transfer learning,” IEEE Trans. on Knowledge and Data Eng., Vol.22, pp. 1345-1359, 2010.
41. [41] J.-K. Zhang, W. Yan, and D.-M. Cui, “Concrete condition assessment using impact-echo method and extreme learning machines,” Sensors, Vol.16, pp. 447-463, 2016.
42. [42] C. Hsiao et al., “Detecting flaws in concrete blocks using the impact-echo method,” NDT & E Int., Vol.41, pp. 98-107, 2008.
43. [43] F. Bettayeb, T. Rachedi, and H. Benbartaoui, “An improved automated ultrasonic NDE system by wavelet and neuron networks,” Ultrasonics, Vol.42, pp. 853-858, 2004.
44. [44] T. D’orazio et al., “Automatic ultrasonic inspection for internal defect detection in composite materials,” NDT & E Int., Vol.41, pp. 145-154, 2008.
45. [45] A. Bernieri et al., “An SVM approach to crack shape reconstruction in eddy current testing,” Proc. of the IEEE Instrumentation and Measurement Tech. Conf., 2006 (IMTC 2006), pp. 2121-2126, 2006.
46. [46] Y. Wang et al., “Detection of line weld defects based on multiple thresholds and support vector machine,” NDT & E Int., Vol.41, pp. 517-524, 2008.
47. [47] E. A. Kholief, S. H. Darwish, and M. N. Fors., “Detection of steel surface defect based on machine learning using deep auto-encoder network,” Proc. of the Int. Conf. on Industrial Engineering and Operations Management, pp. 218-229, 2017.
48. [48] K. Weiss, T. M. Khoshgoftaar, and D. D. Wang, “A survey of transfer learning,” J. of Big Data, Vol.3, 9, 2016.
49. [49] W. Dai et al., “Boosting for Transfer Learning,” Proc. of the 24th Int. Conf. on Machine Learning, pp. 193-200, 2007.
50. [50] H. Daumé III, “Frustratingly easy domain adaptation,” Proc. of the 45th Annual Meeting of the Association of Computational Linguistics, pp. 256-263, 2007.
51. [51] J. Jiang and C. X. Zhai, “A two-stage approach to domain adaptation for statistical classifiers,” Proc. of the 16th ACM Conf. on Information and Knowledge Management, pp. 401-410, 2007.
52. [52] S. Dreiseitl and L. Ohno-Machado, “Logistic regression and artificial neural network classification models: a methodology review,” J. of Biomedical Informatics, Vol.35, pp. 352-359, 2002.
53. [53] G. Stein et al., “Decision tree classifier for network intrusion detection with GA-based feature selection,” Proc. of the 43rd Annual Southeast Regional Conf., Vol.2, pp. 136-141, 2005.
54. [54] A. Alkan and M. Günay, “Identification of EMG signals using discriminant analysis and SVM classifier,” Expert Systems with Applications, Vol.39, pp. 44-47, 2012.
55. [55] A. Khemphila and V. Boonjing, “Comparing performances of logistic regression, decision trees, and neural networks for classifying heart disease patients,” Proc. of the Int. Conf. on Computer Information Systems and Industrial Management Applications (CISIM), pp. 193-198, 2010.
56. [56] S. Mohapatra, D. Patra, and S. Satpathy, “An ensemble classifier system for early diagnosis of acute lymphoblastic leukemia in blood microscopic images,” Neural Computing and Applications, Vol.24, pp. 1887-1904, 2014.
57. [57] J. Zhang, X.-D. Zhang, and S.-W. Ha, “A novel approach using PCA and SVM for face detection,” Proc. of the 4th Int. Conf. on Natural Computation (ICNC’08), Vol.3, pp. 29-33, 2008.