Efficient Industrial Casting Inspection using Deep Transfer Learning
Keywords:
Industry 4.0, Transfer Learning, Casting Defect Detection, VGG-19, Automated Quality Inspection.Abstract
Aluminum alloy castings have become vital elements of numerous contemporary sectors and their structural integrity directly influences the mechanical safety. It is due to this that quality control throughout the manufacturing process is of importance. In this paper, it is proposed to use deep learning to identify the defects in cast products using digital images. A transfer learning approach with a VGG-19 architecture was used to address the difficulties of large-data demands and sophisticated tuning of hyperparameters. The model was also tested on a large data set of 8,012 images including normal and defective samples. We have found that the model has a test accuracy of 99.251 and a recall of 99.265. These results indicate that the presented transfer learning model is a very accurate and robust one which can provide a practical solution to the quality control in real-time within the industrial setting.
References
[1] X. Ji et al., “Filtered selective search and evenly distributed convolutional neural networks for casting defects recognition,” J. Mater. Process. Technol., vol. 292, p. 117064, Jun. 2021, doi: 10.1016/j.jmatprotec.2021.117064.
[2] W. Du, H. Shen, J. Fu, G. Zhang, and Q. He, “Approaches for improvement of the X-ray image defect detection of automobile casting aluminum parts based on deep learning,” NDT E Int., vol. 107, p. 102144, Oct. 2019, doi: 10.1016/j.ndteint.2019.102144.
[3] R. Gupta, V. Anand, S. Gupta, and D. Koundal, “Deep learning model for defect analysis in industry using casting images,” Expert Syst. Appl., vol. 232, no. June, p. 120758, 2023, doi: 10.1016/j.eswa.2023.120758.
[4] Y. Wang, C. Hu, K. Chen, and Z. Yin, “Self-attention guided model for defect detection of aluminium alloy casting on X-ray image,” Comput. Electr. Eng., vol. 88, p. 106821, Dec. 2020, doi: 10.1016/j.compeleceng.2020.106821.
[5] I. Pastor-López, B. Sanz, A. Tellaeche, G. Psaila, J. G. de la Puerta, and P. G. Bringas, “Quality assessment methodology based on machine learning with small datasets: Industrial castings defects,” Neurocomputing, vol. 456, pp. 622–628, Oct. 2021, doi: 10.1016/j.neucom.2020.08.094.
[6] B. Wu et al., “An ameliorated deep dense convolutional neural network for accurate recognition of casting defects in X-ray images,” Knowledge-Based Syst., vol. 226, p. 107096, Aug. 2021, doi: 10.1016/j.knosys.2021.107096.
[7] L. Jiang, Y. Wang, Z. Tang, Y. Miao, and S. Chen, “Casting defect detection in X-ray images using convolutional neural networks and attention-guided data augmentation,” Measurement, vol. 170, p. 108736, Jan. 2021, doi: 10.1016/j.measurement.2020.108736.
[8] H. Han, C. Gao, Y. Zhao, S. Liao, L. Tang, and X. Li, “Polycrystalline silicon wafer defect segmentation based on deep convolutional neural networks,” Pattern Recognit. Lett., vol. 130, pp. 234–241, Feb. 2020, doi: 10.1016/j.patrec.2018.12.013.
[9] “IET Image Processing - 2023 - Wang - Deep learning‐based methods for detecting defects in cast iron parts and surfaces.pdf.” .
[10] X. Wang and X. Yu, “NDT and E International Understanding the effect of transfer learning on the automatic welding defect detection,” NDT E Int., vol. 134, no. May 2022, p. 102784, 2023, doi: 10.1016/j.ndteint.2022.102784.
[11] H. Tanyildiz, “Transfer Learning for Detection of Casting Defects Model In Scope of Industrial 4 . 0 Endüstri 4 . 0 Kapsamında Döküm Hatalarının Tespiti İçin Transfer Öğrenme Modeli,” vol. 12, no. 3, pp. 45–51, 2023.
[12] Z. Song, Y. Yao, and J. Yang, “Casting Defect Detection and Recognition Technology Based,” 2025.
[13] S. Zhang, H. Li, P. Ren, T. Peng, and X. Meng, “OPEN A detection method for small casting defects based on bidirectional feature extraction,” pp. 1–18, 2025.
[14] H. Lin, H. Wu, and C. Lin, “A Deep Transfer Learning-Based Visual Inspection System for Assembly Defects in Similar Types of Manual Tool Products,” 2025.
[15] M. Toğaçar, B. Ergen, and Z. Cömert, “COVID-19 detection using deep learning models to exploit Social Mimic Optimization and structured chest X-ray images using fuzzy color and stacking approaches,” Comput. Biol. Med., vol. 121, p. 103805, Jun. 2020, doi: 10.1016/j.compbiomed.2020.103805.
[16] W. Jin, S. Dong, C. Dong, and X. Ye, “Hybrid ensemble model for differential diagnosis between COVID-19 and common viral pneumonia by chest X-ray radiograph,” Comput. Biol. Med., vol. 131, p. 104252, Apr. 2021, doi: 10.1016/j.compbiomed.2021.104252.
[17] J. Lin, Y. Yao, L. Ma, and Y. Wang, “Detection of a casting defect tracked by deep convolution neural network,” Int. J. Adv. Manuf. Technol., vol. 97, no. 1–4, pp. 573–581, Jul. 2018, doi: 10.1007/s00170-018-1894-0.
[18] T. P. Nguyen, S. Choi, S.-J. Park, S. H. Park, and J. Yoon, “Inspecting Method for Defective Casting Products with Convolutional Neural Network (CNN),” Int. J. Precis. Eng. Manuf. Technol., vol. 8, no. 2, pp. 583–594, Mar. 2021, doi: 10.1007/s40684-020-00197-4.
