{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,19]],"date-time":"2026-05-19T17:13:50Z","timestamp":1779210830122,"version":"3.51.4"},"reference-count":21,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2021,12,28]],"date-time":"2021-12-28T00:00:00Z","timestamp":1640649600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000780","name":"European Union","doi-asserted-by":"publisher","award":["870130"],"award-info":[{"award-number":["870130"]}],"id":[{"id":"10.13039\/501100000780","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Information"],"abstract":"<jats:p>Collecting and labeling of good balanced training data are usually very difficult and challenging under real conditions. In addition to classic modeling methods, Generative Adversarial Networks (GANs) offer a powerful possibility to generate synthetic training data. In this paper, we evaluate the hybrid usage of real-life and generated synthetic training data in different fractions and the effect on model performance. We found that a usage of up to 75% synthetic training data can compensate for both time-consuming and costly manual annotation while the model performance in our Deep Learning (DL) use case stays in the same range compared to a 100% share in hand-annotated real images. Using synthetic training data specifically tailored to induce a balanced dataset, special care can be taken concerning events that happen only on rare occasions and a prompt industrial application of ML models can be executed without too much delay, making these feasible and economically attractive for a wide scope of industrial applications in process and manufacturing industries. Hence, the main outcome of this paper is that our methodology can help to leverage the implementation of many different industrial Machine Learning and Computer Vision applications by making them economically maintainable. It can be concluded that a multitude of industrial ML use cases that require large and balanced training data containing all information that is relevant for the target model can be solved in the future following the findings that are presented in this study.<\/jats:p>","DOI":"10.3390\/info13010009","type":"journal-article","created":{"date-parts":[[2021,12,28]],"date-time":"2021-12-28T01:18:15Z","timestamp":1640654295000},"page":"9","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Impact on Inference Model Performance for ML Tasks Using Real-Life Training Data and Synthetic Training Data from GANs"],"prefix":"10.3390","volume":"13","author":[{"given":"Ulrike","family":"Faltings","sequence":"first","affiliation":[{"name":"SHS\u2014Stahl-Holding-Saar GmbH & Co. KGaA, 66763 Dillingen, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7876-1745","authenticated-orcid":false,"given":"Tobias","family":"Bettinger","sequence":"additional","affiliation":[{"name":"SHS\u2014Stahl-Holding-Saar GmbH & Co. KGaA, 66763 Dillingen, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Swen","family":"Barth","sequence":"additional","affiliation":[{"name":"SHS\u2014Stahl-Holding-Saar GmbH & Co. KGaA, 66763 Dillingen, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michael","family":"Sch\u00e4fer","sequence":"additional","affiliation":[{"name":"SHS\u2014Stahl-Holding-Saar GmbH & Co. KGaA, 66763 Dillingen, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2021,12,28]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Hassanien, A.E., Bhatnagar, R., and Darwish, A. (2021). Deep Learning Techniques: An Overview. Advanced Machine Learning Technologies and Applications, Springer.","DOI":"10.1007\/978-981-15-3383-9"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"4","DOI":"10.3389\/frai.2020.00004","article-title":"An Introductory Review of Deep Learning for Prediction Models With Big Data","volume":"3","author":"Yang","year":"2020","journal-title":"Front. Artif. Intell."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Sun, C., Shrivastava, A., Singh, S., and Gupta, A. (2017). Revisiting Unreasonable Effectiveness of Data in Deep Learning Era. arXiv.","DOI":"10.1109\/ICCV.2017.97"},{"key":"ref_4","first-page":"1767","article-title":"A survey on Image Data Augmentation for Deep Learning","volume":"60","author":"Shorten","year":"2019","journal-title":"J. Big Data"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"249","DOI":"10.1016\/j.neunet.2018.07.011","article-title":"A systematic study of the class imbalance problem in convolutional neural networks","volume":"106","author":"Buda","year":"2018","journal-title":"Neural Netw."},{"key":"ref_6","unstructured":"Perez, L., and Wang, J. (2017). The Effectiveness of Data Augmentation in Image Classification using Deep Learning. arXiv."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Miko\u0142ajczyk, A., and Grochowski, M. (2018, January 9\u201312). Data augmentation for improving deep learning in image classification problem. Proceedings of the 2018 International Interdisciplinary PhD Workshop (IIPhDW), Swinoujscie, Poland.","DOI":"10.1109\/IIPHDW.2018.8388338"},{"key":"ref_8","first-page":"108","article-title":"Physically based synthetic image generation for machine learning: A review of pertinent literature","volume":"Volume 11144","author":"Rosenberger","year":"2019","journal-title":"Proceedings of the Photonics and Education in Measurement Science 2019"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1767","DOI":"10.1007\/s10845-021-01738-7","article-title":"Synthetic image data augmentation for fibre layup inspection processes: Techniques to enhance the data set","volume":"32","author":"Meister","year":"2021","journal-title":"J. Intell. Manuf."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"54207","DOI":"10.1109\/ACCESS.2018.2872025","article-title":"Autoencoder-Combined Generative Adversarial Networks for Synthetic Image Data Generation and Detection of Jellyfish Swarm","volume":"6","author":"Kim","year":"2018","journal-title":"IEEE Access"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Su, Y.H., Jiang, W., Chitrakar, D., Huang, K., Peng, H., and Hannaford, B. (2021). Local Style Preservation in Improved GAN-Driven Synthetic Image Generation for Endoscopic Tool Segmentation. Sensors, 21.","DOI":"10.3390\/s21155163"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1016\/j.promfg.2020.02.084","article-title":"Digital Twin Models in Industrial Operations: A Systematic Literature Review","volume":"42","author":"Melesse","year":"2020","journal-title":"Procedia Manuf."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"346","DOI":"10.1016\/j.jmsy.2020.06.017","article-title":"Review of digital twin about concepts, technologies, and industrial applications","volume":"58","author":"Liu","year":"2021","journal-title":"J. Manuf. Syst."},{"key":"ref_14","unstructured":"Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N., and Weinberger, K.Q. (2014). Generative Adversarial Nets. Advances in Neural Information Processing Systems, Curran Associates, Inc."},{"key":"ref_15","unstructured":"Mirza, M., and Osindero, S. (2014). Conditional Generative Adversarial Nets. arXiv."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Isola, P., Zhu, J.Y., Zhou, T., and Efros, A.A. (2018). Image-to-Image Translation with Conditional Adversarial Networks. arXiv.","DOI":"10.1109\/CVPR.2017.632"},{"key":"ref_17","unstructured":"Cortes, C., Lawrence, N., Lee, D., Sugiyama, M., and Garnett, R. (2015). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. Advances in Neural Information Processing Systems, Curran Associates, Inc."},{"key":"ref_18","unstructured":"(2019, January 15). COCO\u2014Common Objects in Context. Available online: http:\/\/cocodataset.org\/#download."},{"key":"ref_19","unstructured":"(2019, January 15). TensorFlowModelZoo\/Garden. Available online: https:\/\/github.com\/tensorflow\/models."},{"key":"ref_20","unstructured":"(2021, October 13). TensorFlow. Available online: https:\/\/www.tensorflow.org\/."},{"key":"ref_21","unstructured":"Elsayed, G.F., Shankar, S., Cheung, B., Papernot, N., Kurakin, A., Goodfellow, I., and Sohl-Dickstein, J. (2018). Adversarial Examples that Fool both Computer Vision and Time-Limited Humans. arXiv."}],"container-title":["Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2078-2489\/13\/1\/9\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T07:54:31Z","timestamp":1760169271000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2078-2489\/13\/1\/9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,12,28]]},"references-count":21,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2022,1]]}},"alternative-id":["info13010009"],"URL":"https:\/\/doi.org\/10.3390\/info13010009","relation":{},"ISSN":["2078-2489"],"issn-type":[{"value":"2078-2489","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,12,28]]}}}