Abstract
A code-generation and recognition technology that uses a modified ejection system in the diecasting process is presented. To achieve the highest level of quality management, the first requirement in the manufacturing process is to establish a product management system according to the specific product unit. Thus, a method to individually identify each product, such as a barcode or QR code, is required during the production process. Products manufactured in the die-casting process always have ejector pin (EP) marks. Herein, an ejection system was modified to generate a unique code using EP marks. This ejection system has two features: an EP with a modified head to show the direction of rotation, and a function to dependently rotate EPs (five or six EPs) with a constant angle. The EPs are numbered according to the rotation angle. Thus, the EP marks can be viewed as a five- or six-digit code. A program was also developed to individually identify the products by automatically detecting and reading the EPs using deep learning-based object detection and classification technology.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Fraser A, Maltais J, Hartlieb M, et al. Review of technologies for identification of die casting parts. In: Proc. NADCA Die Casting Congress, 2016.
Fraser A, Brochu V, Gingras D, et al. Important considerations for laser marking an identifier on aluminum. Light Metals, 2016: 259–264.
Fraser A and Hartlieb M. Traceability and laser marking of die castings. Die Casting Engineer, 2018: 16–21.
Fanovo. Die Casting. http://fanovo.com/die-casting/Accessed February 4, 2020.
Liu L, Ouyang W L, Wang X G, et al. Deep learning for generic object detection: A survey. International Journal of Computer Vision, 2019, doi: https://doi.org/10.1007/s11263-019-01247-4.
Redmon J, Divvala S, Girshick R, et al. You only look once: unified, real-time object detection. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016: 779–788.
Redmon J and Farhadi A. YOLOv3: An incremental improvement, arXiv:1804.02767v1, 2018.
Ultralytics L L C. YOLOv3 in PyTorch. https://github.com/ultralytics/yolov3/Accessed February 4, 2020.
Girshick R. Fast R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, 2015: 1440–1448.
Girshick R, Donahue J, Darrell T, et al. Region-based convolutional networks for accurate object detection and segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(1): 142–158.
Ren S, He K, Girshick R. et al. Faster R-CNN: Towards realtime object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137–1149.
Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2016, 9905 LNCS: 21–37.
Tzutalin. LabelImg. https://github.com/tzutalin/labelImg/Accessed February 4, 2020.
Krishna S T and Kalluri H K. Deep learning and transfer learning approaches for image classification. International Journal of Recent Technology and Engineering, 2019, 7(5): 427–432.
NVIDIA. What is transfer learning? https://blogs.nvidia.com/blog/2019/02/07/what-is-transfer-learning/ Accessed February 4, 2020.
Tan C Q, Sun F C, Kong T, et al. A survey on deep transfer learning. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2018, 11141 LNCS: 270–279.
Weiss K, Khoshgoftaar T M, and Wang D D. A survey of transfer learning. Journal of Big Data, 2016, 13(1). doi: https://doi.org/10.1186/s40537-016-0043-6.
Lin T Y, Maire M, Belongie S, et al. Microsoft COCO: Common objects in context. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2014, 8693 LNCS (Part 5): 740–755.
COCO. http://cocodataset.org/#home/ Accessed February 4, 2020.
Hui J. mAP (mean Average Precision) for object detection. 2018, https://medium.com/@jonathan_hui/map-mean-average-precision-for-object-detection-45c121a31173/ Accessed 4 February 2020.
McCann S. Average precision. 2020, https://sanchom.wordpress.com/tag/average-precision/ Accessed 4 February 2020.
Padilla R. Metrics for object detection. https://github.com/rafaelpadilla/Object-Detection-Metrics/ Accessed 4 February 2020.
He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016: 770–778.
Acknowledgments
This research was supported by the development project of Industrial and Manufacturing Source Technology of the Korea Institute of Industrial Technology (KITECH), and granted financial resource by the Ministry of Economy and Finance, Republic of Korea (No. EO190031).
Author information
Authors and Affiliations
Corresponding author
Additional information
Chaeho Lim Male, born in 1970, Ph.D., Principle Researcher. Research interests: numerical simulation of the casting process, additive manufacturing of metals, industrial computed tomography technology, and smart manufacturing technology for small & medium factories.
Rights and permissions
About this article
Cite this article
Song, J., Lee, J., Lee, Y.C. et al. Code generation and recognition using a modified ejection system in die-casting process. China Foundry 17, 364–371 (2020). https://doi.org/10.1007/s41230-020-0036-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s41230-020-0036-0