Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network
Source abstract
Abstract
Metal additive manufacturing (AM) is gaining increasing attention from academia and industry due to its unique advantages compared to the traditional manufacturing process. Parts quality inspection is playing a crucial role in the AM industry, which can be adopted for product improvement. However, the traditional inspection process has relied on manual recognition, which could suffer from low efficiency and potential bias. This study presented a convolutional neural network (CNN) approach toward robust AM quality inspection, such as good quality, crack, gas porosity, and lack of fusion. To obtain the appropriate model, experiments were performed on a series of architectures. Moreover, data augmentation was adopted to deal with data scarcity. L2 regularization (weight decay) and dropout were applied to avoid overfitting. The impact of each strategy was evaluated. The final CNN model achieved an accuracy of 92.1%, and it took 8.01 milliseconds to recognize one image. The CNN model presented here can help in automatic defect recognition in the AM industry.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Additive Manufacturing Materials and Processes · Industrial Vision Systems and Defect Detection · Additive Manufacturing and 3D Printing Technologies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA
Funding
National Science Foundation
References (42)
Show source-provided references
1. Machine learning-based image processing for on-line defect recognition in additive manufacturing10.1016/j.cirp.2019.03.021 ↗
2. Functionally graded material of 304L stainless steel and inconel 625 fabricated by directed energy deposition: Characterization and thermodynamic modeling10.1016/j.actamat.2016.02.019 ↗
3. Development and characterization of Ti-6Al-4V to 304L stainless steel gradient components fabricated with laser deposition additive manufacturing10.1016/j.matdes.2016.05.016 ↗
4. Dry sliding wear behavior of laser clad TiVCrAlSi high entropy alloy coatings on Ti-6Al-4V substrate10.1016/j.matdes.2012.04.049 ↗
5. Cui, W., Karnati, S., Zhang, X., Burns, E., and Liou, F. (2019). Fabrication of AlCoCrFeNi high-entropy alloy coating on an AISI 304 substrate via a CoFe2Ni intermediate layer. Entropy, 21.10.3390/e21010002 ↗
6. Influence of microstructure on mechanical properties of laser metal wire-deposited Ti-6Al-4V10.1016/j.msea.2016.07.038 ↗
7. A fractographic study exploring the relationship between the low cycle fatigue and metallurgical properties of laser metal wire deposited Ti-6Al-4V10.1016/j.ijfatigue.2016.02.011 ↗
8. Evaluation of defect density, microstructure, residual stress, elastic modulus, hardness and strength of laser-deposited AISI 4340 steel10.1016/j.actamat.2014.09.028 ↗
9. Powder-based additive manufacturing—A review of types of defects, generation mechanisms, detection, property evaluation and metrology
10. Li, W., Martin, A.J., Kroehler, B., Henderson, A., Huang, T., Watts, J., Hilmas, G.E., and Leu, M.C. (2018, January 13–15). Fabricating Functionally Graded Materials by Ceramic on-Demand Extrusion with Dynamic Mixing. Proceedings of the 29th Annual International Solid Freeform Fabrication Symposium, Austin, TX, USA.
11. A comparative study of laser direct metal deposition characteristics using gas and plasma-atomized Ti-6Al-4V powders10.1016/j.msea.2011.06.074 ↗
12. Characterization of zirconia specimens fabricated by ceramic on-demand extrusion10.1016/j.ceramint.2018.04.008 ↗
13. Advanced steel microstructural classification by deep learning methods10.1038/s41598-018-20037-5 ↗
14. Strip Steel Defect Detection Based on Saliency Map Construction Using Gaussian Pyramid Decomposition10.2355/isijinternational.ISIJINT-2015-041 ↗
15. Vision-based defect detection in laser metal deposition process10.1108/RPJ-04-2012-0036 ↗
16. An End-to-End Steel Strip Surface Defects Recognition System Based on Convolutional Neural Networks10.1002/srin.201600068 ↗
17. American Sign Language alphabet recognition using Convolutional Neural Networks with multiview augmentation and inference fusion10.1016/j.engappai.2018.09.006 ↗
18. Li, K., Wu, Z., Peng, K.C., Ernst, J., and Fu, Y. (2018, January 18–22). Tell Me Where to Look: Guided Attention Inference Network. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00960 ↗
19. Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., and Torralba, A. (2016, January 27–30). Learning Deep Features for Discriminative Localization. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.319 ↗
20. A fast and robust convolutional neural network-based defect detection model in product quality control10.1007/s00170-017-0882-0 ↗
21. Deep learning10.1038/nature14539 ↗
22. Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012). ImageNet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst., 1097–1105.
23. Applying Neural-Network-Based Machine Learning to Additive Manufacturing: Current Applications, Challenges, and Future Perspectives10.1016/j.eng.2019.04.012 ↗
24. Defect classification of laser metal deposition using logistic regression and artificial neural networks for pattern recognition10.1007/s00170-017-0878-9 ↗
25. Porosity prediction: Supervised-learning of thermal history for direct laser deposition10.1016/j.jmsy.2018.04.001 ↗
26. Application of supervised machine learning for defect detection during metallic powder bed fusion additive manufacturing using high resolution imaging
27. Computer Vision and Machine Learning for Autonomous Characterization of AM Powder Feedstocks10.1007/s11837-016-2226-1 ↗
28. Venegas-Barrera, C.S., and Manjarrez, J. (2004, January 11–14). Visual Categorization with Bags of Keypoints. Proceedings of the Workshop on Statistical Learning in Computer Vision, ECCV, Prague, Czech Republic.
29. Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study10.1007/s11263-006-9794-4 ↗
30. Exploring the microstructure manifold: Image texture representations applied to ultrahigh carbon steel microstructures10.1016/j.actamat.2017.05.014 ↗
31. Image driven machine learning methods for microstructure recognition10.1016/j.commatsci.2016.05.034 ↗
32. Microstructure recognition using convolutional neural networks for prediction of ionic conductivity in ceramics10.1016/j.actamat.2017.09.004 ↗
33. Wen, S., Chen, Z., and Li, C. (2018). Vision-based surface inspection system for bearing rollers using convolutional neural networks. Appl. Sci., 8.10.3390/app8122565 ↗
34. Tao, X., Zhang, D., Ma, W., Liu, X., and Xu, D. (2018). Automatic metallic surface defect detection and recognition with convolutional neural networks. Appl. Sci., 8.10.3390/app8091575 ↗
35. Wei, R., and Bi, Y. (2019). Research on Recognition Technology of Aluminum Profile Surface Defects Based on Deep Learning. Materials, 12.10.20944/preprints201904.0322.v1 ↗
36. Zhu, H., Ge, W., and Liu, Z. (2019). Deep Learning-Based Classification of Weld Surface Defects. Appl. Sci., 9.10.3390/app9163312 ↗
37. Liu, Y., Xu, K., and Xu, J. (2019). Periodic Surface Defect Detection in Steel Plates Based on Deep Learning. Appl. Sci., 9.10.3390/app9153127 ↗
38. Additive manufacturing of Ti–6Al–4V parts through laser metal deposition (LMD): Process, microstructure, and mechanical properties10.1016/j.jallcom.2019.04.255 ↗
39. Build Strategy Investigation of Ti-6Al-4V Produced Via a Hybrid Manufacturing Process10.1007/s11837-018-3009-7 ↗
40. Batch normalization: Accelerating deep network training by reducing internal covariate shift
41. Kingma, D.P., and Ba, J.L. (2015, January 7–9). Adam: A method for stochastic optimization. Proceedings of the International Conference on Learning Representations (ICLR), San Diego, CA, USA.
42. Ding, S., Liu, Z., and Li, C. (2011, January 26–28). AdaBoost learning for fabric defect detection based on HOG and SVM. Proceedings of the 2011 International Conference on Multimedia Technology (ICMT), Hangzhou, China.