Towards efficient defect detection in injection molding with transformer models
Andrea PIERESSA, Marco SORGATO, Giovanni LUCCHETTA
Abstract. The startup process in injection molding, traditionally reliant on expert personnel, has become critical due to small-batch production and a shortage of skilled workers. This shift requires efficient quality monitoring, especially for visually demanding plastic products. This study evaluates transformer-based computer vision models like RT-DETR for detecting defects characterized by domain-specific patterns and high intra- and inter-class variability. The research compares general models with specialized models trained for specific component-defect combinations, focusing on key computer vision metrics and computational demands. Injection molding features such as representative dimensions and aspect ratio are analyzed to assess knowledge transferability from pre-trained models to new scenarios. The findings could enhance defect detection accuracy, reduce reliance on experts, and boost production efficiency in small-batch, visually critical manufacturing.
Keywords
Object Recognition, Zero Defect, Injection Molding
Published online 9/10/2025, 8 pages
Copyright © 2025 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA
Citation: Andrea PIERESSA, Marco SORGATO, Giovanni LUCCHETTA, Towards efficient defect detection in injection molding with transformer models, Materials Research Proceedings, Vol. 57, pp 600-607, 2025
DOI: https://doi.org/10.21741/9781644903735-70
The article was published as article 70 of the book Italian Manufacturing Association Conference
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
References
[1] J.-B. E. M. Steenkamp, «Conceptual model of the quality perception process», Journal of Business Research, vol. 21, fasc. 4, pp. 309–333, dic. 1990. https://doi.org/10.1016/0148-2963(90)90019-A
[2] Z. Chen e L. Turng, «A review of current developments in process and quality control for injection molding», Adv Polym Technol, vol. 24, fasc. 3, pp. 165–182, set. 2005. https://doi.org/10.1002/adv.20046
[3] K.-M. Tsai, C.-Y. Hsieh, e W.-C. Lo, «A study of the effects of process parameters for injection molding on surface quality of optical lenses», Journal of Materials Processing Technology, vol. 209, fasc. 7, pp. 3469–3477, apr. 2009. https://doi.org/10.1016/j.jmatprotec.2008.08.006
[4] J. Gim e L.-S. Turng, «A review of current advancements in high surface quality injection molding: Measurement, influencing factors, prediction, and control», Polymer Testing, vol. 115, p. 107718, nov. 2022. https://doi.org/10.1016/j.polymertesting.2022.107718
[5] C.-C. An e R.-H. Chen, «The experimental study on the defects occurrence of SL mold in injection molding», Journal of Materials Processing Technology, vol. 201, fasc. 1–3, pp. 706–709, mag. 2008. https://doi.org/10.1016/j.jmatprotec.2007.11.179
[6] R. Kerkstra e S. Brammer, Injection Molding Advanced Troubleshooting Guide. München: Carl Hanser Verlag GmbH & Co. KG, 2018. https://doi.org/10.3139/9781569906460
[7] G. Baruffa, A. Pieressa, M. Sorgato, e G. Lucchetta, «Transfer Learning-Based Artificial Neural Network for Predicting Weld Line Occurrence through Process Simulations and Molding Trials», JMMP, vol. 8, fasc. 3, p. 98, mag. 2024. https://doi.org/10.3390/jmmp8030098
[8] R. Zhang, Y. Li, e Z. Guan, «Research on injection molded parts defect detection algorithm based on multiplicative feature fusion and improved attention mechanism», Sci Rep, vol. 14, fasc. 1, p. 30864, dic. 2024. https://doi.org/10.1038/s41598-024-81430-x
[9] J. Gim, H. Yang, e L.-S. Turng, «Transfer learning of machine learning models for multi-objective process optimization of a transferred mold to ensure efficient and robust injection molding of high surface quality parts», Journal of Manufacturing Processes, vol. 87, pp. 11–24, feb. 2023. https://doi.org/10.1016/j.jmapro.2022.12.055
[10] J. Liu, F. Guo, H. Gao, M. Li, Y. Zhang, e H. Zhou, «Defect detection of injection molding products on small datasets using transfer learning», Journal of Manufacturing Processes, vol. 70, pp. 400–413, ott. 2021. https://doi.org/10.1016/j.jmapro.2021.08.034
[11] A. Pieressa, G. Baruffa, M. Sorgato, e G. Lucchetta, «Enhancing weld line visibility prediction in injection molding using physics-informed neural networks», J Intell Manuf, lug. 2024. https://doi.org/10.1007/s10845-024-02460-w
[12] G. A. Pereira e M. Hussain, «A Review of Transformer-Based Models for Computer Vision Tasks: Capturing Global Context and Spatial Relationships», 27 agosto 2024, arXiv: arXiv:2408.15178. https://doi.org/10.48550/arXiv.2408.15178
[13] A. Dosovitskiy et al., «An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale», 3 giugno 2021 arXiv: arXiv:2010.11929. https://doi.org/10.48550/arXiv.2010.11929.
[14] S. Khan, M. Naseer, M. Hayat, S. W. Zamir, F. S. Khan, e M. Shah, «Transformers in Vision: A Survey», ACM Comput. Surv., vol. 54, fasc. 10s, set. 2022. https://doi.org/10.1145/3505244
[15] N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, e S. Zagoruyko, «End-to-End Object Detection with Transformers», 28 maggio 2020, arXiv: arXiv:2005.12872. https://doi.org/10.48550/arXiv.2005.12872
[16] Y. Zhao et al., «DETRs Beat YOLOs on Real-time Object Detection», 3 aprile 2024, arXiv: arXiv:2304.08069. https://doi.org/10.48550/arXiv.2304.08069
