{"id":119522,"key":"Synthetic-to-real_Composite_Semantic_Segmentation_in_Additive_Manufacturing","title":"Synthetic-to-real Composite Semantic Segmentation in Additive Manufacturing","latest":{"id":1152946,"timestamp":"2025-03-06T13:04:10Z"},"content_model":"wikitext","license":{"url":"https://www.appropedia.org/Appropedia:Copyrights","title":"CC-BY-SA-4.0"},"source":"{{FAST notice}}\n\n[[File:Graphical abstract-synth2real.jpg|thumb]]\n\n{{Publication data\n| type = Paper\n| cite-as = Petsiuk A, Singh H, Dadhwal H, Pearce JM. Synthetic-to-Real Composite Semantic Segmentation in Additive Manufacturing. ''Journal of Manufacturing and Materials Processing''. 2024; 8(2):66. https://doi.org/10.3390/jmmp8020066 [https://www.academia.edu/116791164/Synthetic_to_Real_Composite_Semantic_Segmentation_in_Additive_Manufacturing Academia open access] [https://arxiv.org/abs/2210.07466 ArXiv]\n}}\n\n{{Project data\n| authors = Aliaksei L. Petsiuk, H. Singh, H. Dadhwal, User:J.M.Pearce\n| status = Designed, Modelled, Prototyped, Verified\n| verified-by = FAST\n| made = Yes\n| years = 2024\n| type = 3D Printing, AI, computer vision, machine learning\n| location = London, ON\n}}\n\nThe application of computer vision and machine learning methods for semantic segmentation of the structural elements of 3D-printed products in the field of additive manufacturing (AM) can improve real-time failure analysis systems and potentially reduce the number of defects by providing additional tools for in situ corrections. This work demonstrates the possibilities of using physics-based rendering for labeled image dataset generation, as well as image-to-image style transfer capabilities to improve the accuracy of real image segmentation for AM systems. Multi-class semantic segmentation experiments were carried out based on the U-Net model and the cycle generative adversarial network. The test results demonstrated the capacity of this method to detect such structural elements of 3D-printed parts as a top (last printed) layer, infill, shell, and support. A basis for further segmentation system enhancement by utilizing image-to-image style transfer and domain adaptation technologies was also considered. The results indicate that using style transfer as a precursor to domain adaptation can improve real 3D printing image segmentation in situations where a model trained on synthetic data is the only tool available. The mean intersection over union (mIoU) scores for synthetic test datasets included 94.90% for the entire 3D-printed part, 73.33% for the top layer, 78.93% for the infill, 55.31% for the shell, and 69.45% for supports.\n\n* Free and open source code: https://osf.io/h8r45\n* [[Blender]]\n\n{{Pearce publications notice}}\n\n== Keywords ==\n\n[[3-D printing]], [[additive manufacturing]], g-code segmentation, sim-to-real, semantic segmentation, synthetic data, machine learning, open source software, [[open-source hardware]], [[RepRap]], computer vision, quality assurance, real-time monitoring, anomaly detection; Blender, synthetic images\n\n== See also ==\n\n{{FAST-CV}}\n\n* [[Integrated Voltage—Current Monitoring and Control of Gas Metal Arc Weld Magnetic Ball-Jointed Open Source 3-D Printer]]\n* [[Low-cost Open-Source Voltage and Current Monitor for Gas Metal Arc Weld 3-D Printing]]\n* [[Slicer and process improvements for open-source GMAW-based metal 3-D printing]]\n* [[Open-source Lab]]\n* [[Open source 3-D printing of OSAT]]\n\n{{MOST-RepRap}}\n\n{{Page data\n| part-of = FAST Completed\n| keywords = [[3D printing]], [[additive manufacturing]], g-code segmentation, sim-to-real, semantic segmentation, synthetic data, machine learning, open source software, [[open-source hardware]], [[RepRap]], computer vision, quality assurance, real-time monitoring, anomaly detection; Blender, synthetic images\n| sdg = SDG09 Industry innovation and infrastructure\n| published = 2024\n| organizations = Free Appropriate Sustainable Technology, Western\n| license = CC-BY-SA-4.0\n| language = en\n| authors = Aliaksei L. Petsiuk, User:J.M.Pearce\n| title-tag = In-Situ Layer-Wise 3D Printing Anomaly Detection\n}}\n\n[[Category:3D printing]]\n[[Category:AI]]"}