Self-Supervised Learning Methods for Limited Labelled Data in Manufacturing Quality Control
Keywords:
Laser-based Directed Energy Deposition (L-DED), X-Ray Images, Solid Foundation, Manufacturing Products, Convolutional Neural Networks, Industrial Dataset, Deep Learning Techniques.Abstract
Successful deep learning model deployment often depends on the quantity, quality, and accessibility of annotated data, as the use of deep learning methods in industrial applications expands at an accelerating rate and scale. The issues of effective data labelling and annotation verification in a human-in-the-loop scenario are addressed in this work. The Laser-based Directed Energy Deposition (L-DED) procedure is the subject of this work, which makes use of embedded vision systems to record crucial melt pool properties for ongoing observation. In order to provide in-situ monitoring without ground truth information, two self-learning frameworks based on Transformer architecture and Convolutional Neural Networks are deployed to analyse zone pictures from various DED process regimes. Although they need explicit human supervision, deep convolutional neural networks have recently shown respectable improvement in learning spatial patterns in WBMs. Furthermore, the RGB pictures that make up the majority of these datasets vary greatly from X-ray images. To overcome this drawback, our study suggests an approach that uses X-ray imaging and domain-specific self-supervised pretraining methods to enhance the ability to identify defects in manufactured goods. To improve feature extraction from manufacturing photos, we use SimSiam and SimMIM, two pretraining techniques. An industrial dataset of 27,901 unlabelled X-ray pictures from a manufacturing production line is used for the pretraining phase. Furthermore, we highlight how the models pretrained using X-ray pictures have improved their capacity to identify important flaws, which is essential for maintaining safety in industrial environments. Significant proof of the advantages of self-supervised learning in manufacturing defect identification is provided by our study, laying the groundwork for future investigations and useful applications in industrial quality control.
Downloads
References
Wenbin Cai, Ya Zhang, Siyuan Zhou, Wenquan Wang, Chris Ding, and Xiao Gu. Active learning for support vector machines with maximum model change. In Machine Learning and Knowledge Discovery in Databases, pages 211–226. Springer, 2014.
Yi Yang, Zhigang Ma, Feiping Nie, Xiaojun Chang, and Alexander G Hauptmann. Multi-class active learning by uncertainty sampling with diversity maximization. IJCV, 113(2):113–127, 2015.
Yuhong Guo. Active instance sampling via matrix partition. In NIPS, pages 1–9, 2010.
Suyog Dutt Jain and Kristen Grauman. Active image segmentation propagation. In CVPR, pages 2864– 2873, 2016.
Ozan Sener and Silvio Savarese. Active learning for convolutional neural networks: A core-set approach. In ICLR, 2018.
Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. A simple framework for contrastive learning of visual representations. In International conference on machine learning, pages 1597–1607. PMLR, 2020.
Kaiming He, Haoqi Fan, Yuxin Wu, Saining Xie, and Ross Girshick. Momentum contrast for unsupervised visual representation learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9729–9738, 2020.
Xinlei Chen, Haoqi Fan, Ross Girshick, and Kaiming He. Improved baselines with momentum contrastive learning. arXiv preprint arXiv:2003.04297, 2020.
Jean-Bastien Grill, Florian Strub, Florent Altché, Corentin Tallec, Pierre Richemond, Elena Buchatskaya, Carl Doersch, Bernardo Avila Pires, Zhaohan Guo, Mohammad Gheshlaghi Azar, Bilal Piot, koray kavukcuoglu, Remi Munos, and Michal Valko. Bootstrap your own latent - a new approach to self-supervised learning. In H. Larochelle, M. Ranzato, R. Hadsell, M. F. Balcan, and H. Lin, editors, Advances in Neural Information Processing Systems, volume 33, pages 21271–21284. Curran Associates, Inc., 2020.
Mathilde Caron, Ishan Misra, Julien Mairal, Priya Goyal, Piotr Bojanowski, and Armand Joulin. Unsupervised learning of visual features by contrasting cluster assignments. arXiv preprint arXiv:2006.09882, 2020.
Xinlei Chen and Kaiming He. Exploring simple siamese representation learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 15750–15758, 2021.
Antti Rasmus, Mathias Berglund, Mikko Honkala, Harri Valpola, and Tapani Raiko. Semi-supervised learning with ladder networks. In C. Cortes, N. Lawrence, D. Lee, M. Sugiyama, and R. Garnett, editors, Advances in Neural Information Processing Systems, volume 28. Curran Associates, Inc., 2015.
Mery, D.; Arteta, C. Automatic Defect Recognition in X-ray Testing Using Computer Vision. In Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, USA, 24–31 March 2017; pp. 1026–1035.
Li, X.; Tso, S.K.; Guan, X.P.; Huang, Q. Improving Automatic Detection of Defects in Castings by Applying Wavelet Technique. IEEE Trans. Ind. Electron. 2006, 53, 1927–1934.
Mery, D. Automated Radioscopic Inspection of Aluminum Die Castings. Mater. Eval. 2006, 65, 643–647.
Tsai, D.M.; Huang, T.Y. Automated Surface Inspection for Statistical Textures. Image Vis. Comput. 2003, 21, 307–323.
Zhao, X.; He, Z.; Zhang, S.; Liang, D. A Sparse-Representation-Based Robust Inspection System for Hidden Defects Classification in Casting Components. Neurocomputing 2015, 153, 1–10.
Du, W.; Shen, H.; Fu, J.; Zhang, G.; He, Q. Approaches for Improvement of the X-ray Image Defect Detection of Automobile Casting Aluminium Parts Based on Deep Learning. NDT Int. 2019, 107, 102144.
Mery, D. Aluminum Casting Inspection Using Deep Object Detection Methods and Simulated Ellipsoidal Defects. Mach. Vis. Appl. 2021, 32, 72.
Kim, H., Cong, W., Zhang, H.-C., & Liu, Z. (2017). Laser engineered net shaping of nickel-based superalloy Inconel 718 powders onto AISI 4140 alloy steel substrates: Interface bond and fracture failure mechanism. Materials, 10(4), 341.
Lei, J. B., Wang, Z., & Wang, Y. S. (2012). Measurement on temperature distribution of metal powder stream in laser fabricating. Applied Mechanics and Materials.
Li, X., Siahpour, S., Lee, J., Wang, Y., & Shi, J. (2020). Deep learningbased intelligent process monitoring of directed energy deposition in additive manufacturing with thermal images. Procedia Manufacturing, 48, 643–649.
Liu, X., Zhang, F., Hou, Z., Mian, L., Wang, Z., Zhang, J., & Tang, J. (2021b). Self-supervised learning: Generative or contrastive. IEEE Transactions on Knowledge and Data Engineering.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
