Automatic detection and characterization of porosities in cross-section images of metal parts produced by binder jetting using machine learning and image augmentation
Pages 1281 - 1303
Abstract
In binder jetting followed by sintering, the porosity characterization is critical to understand how the process affects the structure of the printed parts. Image-based porosity detection methods are widely used but the current solutions are limited to specific materials and conditions and require manual tuning that precludes real-time porosity detection. The application of machine learning for automating porosity detection has been also limited to specific materials and conditions and requires a large training dataset for successful implementation. However, large datasets are difficult to acquire experimentally in binder jetting due to prohibited material costs and experiment time. To bridge the knowledge gap, this paper investigates the application of machine learning on automated porosity detection using a small dataset consisting of highly varied cross-section images of metal parts produced by binder jetting followed by sintering. Stylegan3, a type of generative adversarial network, is used to increase the number of training images by image augmentation, and YOLOv5, a convolutional neural network specialized for object detection, is used to detect porosities. The resulting model achieves an F1 score of 88% and detection time of 3–15 ms per image. Generalized porosity detection is also assessed on a set of images containing highly varied materials, resolutions, magnifications, and pore densities. Furthermore, morphological information of the classified porosities such as the distribution of their orientations are automatically extracted using image processing algorithms.
References
[1]
Abdelrahman M, Reutzel EW, Nassar AR, and Starr TL Flaw detection in powder bed fusion using optical imaging Additive Manufacturing 2017 15 1-11
[2]
Aminzadeh M and Kurfess TR Online quality inspection using Bayesian classification in powder-bed additive manufacturing from high-resolution visual camera images Journal of Intelligent Manufacturing 2019 30 6 2505-2523
[3]
An, C., Wen, X., Wang, J., & Wu, B. (2016). GAP/DNTF based PBX explosives: A novel formula used in small sized explosive circuits. Central European Journal of Energetic Materials.
[4]
Andreeva R, Stoyanova E, Tsanev A, and Stoychev D Corrosion behavior of anodically formed oxide films on aluminum, sealed in cerium-ions containing solutions Bulgarian Chemical Communications 2016 48 96-102
[5]
Antunes RA, Ichikawa RU, Martinez LG, and Costa I Characterization of corrosion products on carbon steel exposed to natural weathering and to accelerated corrosion tests International Journal of Corrosion 2014
[6]
ASTM. (2012). F2792-12a, Standard Terminology for Additive Manufacturing Technologies, (Withdrawn 2015). ASTM International.
[7]
Baek JY, Xing ZC, Kwak G, Yoon KB, Park SY, Park LS, and Kang IK Fabrication and characterization of collagen-immobilized porous PHBV/HA nanocomposite scaffolds for bone tissue engineering Journal of Nanomaterials 2012
[8]
Bai Y and Williams CB An exploration of binder jetting of copper Rapid Prototyping Journal 2015
[9]
Bergmann C, Lindner M, Zhang W, Koczur K, Kirsten A, Telle R, and Fischer H 3D printing of bone substitute implants using calcium phosphate and bioactive glasses Journal of the European Ceramic Society 2010
[10]
Bochkovskiy, A., Wang, C.-Y., & Liao, H.-Y. M. (2020). YOLOv4: optimal speed and accuracy of object detection. arXiv Preprint.
[11]
Chen F, Qiu B, Ye Y, Lv Y, Shangguan Y, Song Y, and Zheng Q Reconstruction of core-shell dispersed particles in impact polypropylene copolymer during extrusion Chinese Journal of Polymer Science (English Edition) 2015
[12]
Chen H and Zhao YF Process parameters optimization for improving surface quality and manufacturing accuracy of binder jetting additive manufacturing process Rapid Prototyping Journal 2016 22 3 527-538
[13]
Cheng JCP and Wang M Automated detection of sewer pipe defects in closed-circuit television images using deep learning techniques Automation in Construction 2018
[14]
Chia, H. (2012). Southern California Society for microscopy and microanalysis. Retrieved June 16, 2022, from http://www.scsmm.org/gallery.html
[15]
Community, O. (2010). The OpenCV reference manual. October.
[16]
Cui X, Shi J, Ye Z, Zhang Z, Xu B, and Chen S Layer-by-layer assembly and photocatalytic activity of titania nanosheets on coal fly ash microspheres International Journal of Photoenergy 2014
[17]
Da’as EH, Irvine JTS, Traversa E, and Boulfrad S Controllable impregnation via Inkjet printing for the fabrication of solid oxide cell air electrodes ECS Transactions 2013
[18]
Do T, Kwon P, and Shin CS Process development toward full-density stainless steel parts with binder jetting printing International Journal of Machine Tools and Manufacture 2017 121 50-60
[19]
Do T, Shin CS, Stetsko D, Vanconant G, Vartanian A, Pei S, and Kwon P Improving structural integrity with boron-based additives for 3D printed 420 stainless steel Procedia Manufacturing 2015
[20]
Dojcinovic M, Eric O, Rajnovic D, Sidanin L, and Balos S The morphology of ductile cast iron surface damaged by cavitation Metallurgical & Materials Engineering 2012 18 3 165-176
[21]
Dollár P and Zitnick CL Fast edge detection using structured forests IEEE Transactions on Pattern Analysis and Machine Intelligence 2015
[22]
Doyle M, Agarwal K, Sealy W, and Schull K Effect of layer thickness and orientation on mechanical behavior of binder jet stainless steel 420 + bronze parts Procedia Manufacturing 2015
[23]
Enneti RK, Prough KC, Wolfe TA, Klein A, Studley N, and Trasorras JL Sintering of WC-12%Co processed by binder jet 3D printing (BJ3DP) technology International Journal of Refractory Metals and Hard Materials 2018
[24]
Fisher Scientific. (2022). MilliporeSigmaTM ExpressTM PLUS membrane filters. Fisher Scientific Product Catalog. Retrieved November 20, 2022, from https://www.fishersci.com/shop/products/emd-millipore-express-plus-membrane-filters-10/p-3303398
[25]
García-Moreno AI, Alvarado-Orozco JM, Ibarra-Medina J, and Martínez-Franco E Image-based porosity classification in Al-alloys by laser metal deposition using random forests International Journal of Advanced Manufacturing Technology 2020
[26]
Gaytan SM, Cadena MA, Karim H, Delfin D, Lin Y, Espalin D, MacDonald E, and Wicker RB Fabrication of barium titanate by binder jetting additive manufacturing technology Ceramics International 2015
[27]
Gibson I, Rosen D, and Stucker B Additive manufacturing technologies: 3D printing and direct digital manufacturing 2015 Springer
[28]
Gobert C, Reutzel EW, Petrich J, Nassar AR, and Phoha S Application of supervised machine learning for defect detection during metallic powder bed fusion additive manufacturing using high resolution imaging Additive Manufacturing 2018 21 517-528
[29]
Gonzalez JA, Mireles J, Lin Y, and Wicker RB Characterization of ceramic components fabricated using binder jetting additive manufacturing technology Ceramics International 2016
[30]
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE computer society conference on computer vision and pattern recognition (Vol. 2016-December).
[31]
Heusel, M., Ramsauer, H., Unterthiner, T., Nessler, B., & Hochreiter, S. (2017). GANs trained by a two time-scale update rule converge to a local Nash equilibrium. In Advances in neural information processing systems (Vol. 2017-December).
[32]
Hong D, Chou DT, Velikokhatnyi OI, Roy A, Lee B, Swink I, Issaev I, Kuhn H, and Kumta PN Binder-jetting 3D printing and alloy development of new biodegradable Fe–Mn–Ca/Mg alloys Acta Biomaterialia 2016
[33]
Huang H, Zhou H, Yang X, Zhang L, Qi L, and Zang AY Faster R-CNN for marine organisms detection and recognition using data augmentation Neurocomputing 2019
[34]
Hussain M, Bird JJ, and Faria DR A study on CNN transfer learning for image classification Advances in Intelligent Systems and Computing 2019
[35]
Imani F, Gaikwad A, Montazeri M, Rao P, Yang H, and Reutzel E Process mapping and in-process monitoring of porosity in laser powder bed fusion using layerwise optical imaging Journal of Manufacturing Science and Engineering, Transactions of the ASME 2018
[36]
Islam MN and Sacks S An experimental investigation into the dimensional error of powder-binder three-dimensional printing International Journal of Advanced Manufacturing Technology 2016
[37]
Jamion, N. A. B., Hafiff, N. H. B. A., Halim, N. H. A., Izzaddin, S. A., Ghazali, S. M., & Sapari, J. M. (2017). Preparation of date seed activation for surfactant recovery. Malaysian Journal of Analytical Science.
[38]
Jocher, G., Stoken, A., Borovec, J., Christopher, S., Changyu, L., Hogan, A., Laughing, A., Diaconu, L., Poznanski, J., Yu, L., Rai, R., Ferriday, R., Sullivan, T., Claramunt, E. R. (2020). ultralytics/yolov5: v2.0. GitHub.
[39]
Jocher, G., Stoken, A., Borovec, J., Chaurasia, A., Changyu, L., Laughing, A., Hogan, A., Hajek, J., Diaconu, L., Marc, Y. (2021). ultralytics/yolov5: v5.0—YOLOv5-P6 1280 models, AWS, Supervise.ly and YouTube integrations. Zenodo.
[40]
Karras, T., Aittala, M., Laine, S., Härkönen, E., Hellsten, J., Lehtinen, J., & Aila, T. (2021). Alias-free generative adversarial networks. arXiv Preprint.
[41]
Lanzetta, M., & Sachs, E. M. (2001). The line formation with alumina powders in drop on demand three dimensional printing. In Proceedings of the 1st international seminar on: Progress in innovative manufacturing engineering (PRIME 2001).
[42]
Lee G, Carrillo M, McKittrick J, Martin DG, and Olevsky EA Fabrication of ceramic bone scaffolds by solvent jetting 3D printing and sintering: Towards load-bearing applications Additive Manufacturing 2020
[43]
Lei HW, Wang B, Wu HH, and Wang AH Defect detection for polymeric polarizer based on faster R-CNN Journal of Information Hiding and Multimedia Signal Processing 2018 9 6 1414-1420
[44]
Li S, Gu X, Xu X, Xu D, Zhang T, Liu Z, and Dong Q Detection of concealed cracks from ground penetrating radar images based on deep learning algorithm Construction and Building Materials 2021
[45]
Li Y, Yang C, Zhao H, Qu S, Li X, and Li Y New developments of ti-based alloys for biomedical applications Materials 2014
[46]
Liu Y, Cen C, Che Y, Ke R, Ma Y, and Ma Y Detection of maize tassels from UAV RGB imagery with faster R-CNN Remote Sensing 2020
[47]
Manière C and Olevsky EA Porosity dependence of powder compaction constitutive parameters: Determination based on spark plasma sintering tests Scripta Materialia 2017
[48]
Milewski, J. O. (2017). Additive manufacturing metal, the art of the possible. In Springer series in materials science. Springer.
[49]
Miyanaji H, Zhang S, Lassell A, Zandinejad A, and Yang L Process development of porcelain ceramic material with binder jetting process for dental applications JOM Journal of the Minerals Metals and Materials Society 2016
[50]
Moon J, Caballero AC, Hozer L, Chiang YM, and Cima MJ Fabrication of functionally graded reaction infiltrated SiC-Si composite by three-dimensional printing (3DP™) process Materials Science and Engineering A 2001
[51]
Mostafaei A, Stevens EL, Hughes ET, Biery SD, Hilla C, and Chmielus M Powder bed binder jet printed alloy 625: Densification, microstructure and mechanical properties Materials and Design 2016
[52]
Olevsky EA Theory of sintering: From discrete to continuum Materials Science and Engineering R: Reports 1998 23 2 41-100
[53]
Ozols, A., Thern, G., Rozenberg, S., Barreiro, M., & Marajofsky, A. (2004). Moldeo de Acero AISI 316 por Colada Gel. CONGRESO CONAMET/SAM.
[54]
Pinkney, J. (2022, June 12). Awesome pretrained StyleGAN3. GitHub. Retrieved June 14, 2022, from https://github.com/justinpinkney/awesome-pretrained-stylegan3
[55]
Ren S, He K, Girshick R, and Sun J Faster R-CNN: Towards real-time object detection with region proposal networks IEEE Transactions on Pattern Analysis and Machine Intelligence 2017
[56]
Sachs E, Cima M, and Cornie J Three-dimensional printing: Rapid tooling and prototypes directly from a CAD model CIRP Annals—Manufacturing Technology 1990
[57]
Sachs E, Cima M, Williams P, Brancazio D, and Cornie J Three dimensional printing: Rapid tooling and prototypes directly from a CAD model Journal of Manufacturing Science and Engineering, Transactions of the ASME. 1992
[58]
Sasaki, Y. (2007). The truth of the F-measure. Teach Tutor mater.
[59]
Satterlee N, Torresani E, Olevsky E, and Kang JS Comparison of machine learning methods for automatic classification of porosities in powder-based additive manufactured metal parts The International Journal of Advanced Manufacturing Technology 2022 120 9 6761-6776
[60]
Scheiba, F. (2021). Carbon based materials for electrochemical energy storage. Institute for Applied Materials Energy Storage Systems. Retrieved June 16, 2022, from https://www.iam.kit.edu/ess/english/667.php
[61]
Scime L and Beuth J A multi-scale convolutional neural network for autonomous anomaly detection and classification in a laser powder bed fusion additive manufacturing process Additive Manufacturing 2018 24 273-286
[62]
Shrestha S and Manogharan G Optimization of binder jetting using Taguchi method JOM Journal of the Minerals Metals and Materials Society 2017 69 3 491-497
[63]
Snelling D, Li Q, Meisel N, Williams CB, Batra RC, and Druschitz AP Lightweight metal cellular structures fabricated via 3D printing of sand cast molds Advanced Engineering Materials 2015
[64]
Srinivas M and Babu BS A critical review on recent research methodologies in additive manufacturing Materials Today: Proceedings 2017
[65]
Sun X, Gu J, Huang R, Zou R, and Palomares BG Surface defects recognition of wheel hub based on improved faster R-CNN Electronics (Switzerland) 2019
[66]
Tang Y, Zhou Y, Hoff T, Garon M, and Zhao YF Elastic modulus of 316 stainless steel lattice structure fabricated via binder jetting process Materials Science and Technology (United Kingdom) 2016
[67]
Wang Y and Zhao YF Investigation of sintering shrinkage in binder jetting additive manufacturing process Procedia Manufacturing 2017
[68]
Winkel A, Meszaros R, Reinsch S, Müller R, Travitzky N, Fey T, Greil P, and Wondraczek L Sintering of 3D-printed glass/HAp composites Journal of the American Ceramic Society 2012
[69]
Wu W, Yin Y, Wang X, and Xu D Face detection with different scales based on faster R-CNN IEEE Transactions on Cybernetics 2019
[70]
Wusatowska-Sarnek AM, Ghosh G, Olson GB, Blackburn MJ, and Aindow M Characterization of the microstructure and phase equilibria calculations for the powder metallurgy superalloy IN100 Journal of Materials Research 2003
[71]
Xia, W., Zhang, Y., Yang, Y., Xue, J.-H., Zhou, B., & Yang, M.-H. (2021). GAN inversion: A survey. arXiv Preprint.
[72]
Yao J, Qi J, Zhang J, Shao H, Yang J, and Li X A real-time detection algorithm for kiwifruit defects based on yolov5 Electronics (Switzerland) 2021
[73]
Yao-jian, R., Zhi, S., Jie, T., & Wei, L. (2009). Study on electrolytic aluminium carbon anode preparation with calcined anthracite. In Procedia Earth and planetary science (Vol. 1).
[74]
Yoo, J., Cima, M., Sachs, E., & Suresh, S. (2008). Fabrication and microstructural control of advanced ceramic components by three dimensional printing. In Proceedings of the 19th annual conference on composites, advanced ceramics, materials, and structures—B: Ceramic engineering and science proceedings.
[75]
Zhang B, Liu S, and Shin YC In-Process monitoring of porosity during laser additive manufacturing process Additive Manufacturing 2019
[76]
Zhang, S., Miyanaji, H., Yang, L., Zandinejad, A. A., Dilip, J. J. S., & Stucker, B. (2014). An experimental study of ceramic dental porcelain materials using a 3D print (3DP) process. In 25th Annual international solid freeform fabrication symposium; An additive manufacturing conference, SFF 2014.
[77]
Zhang X, Hao Y, Shangguan H, Zhang P, and Wang A Detection of surface defects on solar cells by fusing multi-channel convolution neural networks Infrared Physics and Technology 2020
[78]
Zhao J, Zhang X, Yan J, Qiu X, Yao X, Tian Y, Zhu Y, and Cao W A wheat spike detection method in UAV images based on improved YOLOv5 Remote Sensing 2021 13 16 3095
[79]
Zhou Y, Tang Y, Hoff T, Garon M, and Zhao FY The verification of the mechanical properties of binder jetting manufactured parts by instrumented indentation testing Procedia Manufacturing 2015
[80]
Zhu, X., Lyu, S., Wang, X., & Zhao, Q. (2021). TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios. In Proceedings of the IEEE international conference on computer vision (Vol. 2021-October).
Recommendations
Development of micro-stereolithography technology using metal powder
Micro-stereolithography is a new technology that can be used to fabricate free-form 3D microstructures by dividing a desired shape into many slices of a given horizontal thickness, hardening each slice with a focused laser beam, and then stacking them ...
Electrochemical migration of directly printed Ag electrodes using Ag paste with epoxy binder
In this study, we investigated the phenomena of electrochemical migration (ECM) on a silicon (Si) substrate with directly printed Ag paste with epoxy binder (7wt.%) using a screen-printing method. The microstructure and electrical characteristics of the ...
Ceramic micro parts produced by micro injection molding: latest developments
Powder injection molding is a preferred technology for the production of micro parts or microstructured parts. Derived from the well known thermoplastic injection molding technique it is suitable for a large-scale production of ceramic and metallic ...
Comments
Information & Contributors
Information
Published In
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023. 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.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 01 April 2023
Accepted: 22 February 2023
Received: 08 July 2022
Author Tags
Qualifiers
- Research-article
Funding Sources
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Other Metrics
Citations
View Options
View options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in