This study introduces a novel deep learning-based approach for the automated and accurate detection of surface defects in piston rings, utilizing convolutional neural networks (CNNs). A high-quality dataset of defective and non-defective piston ring images was collected and meticulously annotated. We trained a U-Net model on this dataset, which effectively segments defective regions, achieving an impressive Intersection over Union (IoU) score of 84% and a loss of 0.15. These results demonstrate the model’s high precision in identifying various defect types. A comparative analysis with traditional defect detection methods underscores the superiority of our deep learning approach in both accuracy and processing speed. Additionally, the experimental findings align closely with theoretical predictions from prior research, further validating the proposed model. This research represents a significant advancement in automotive quality control, offering the potential to reduce manufacturing costs associated with defective components.
[1] Liang JC. An integrated product development process in the automotive industry. International Journal of Product Development. 2009 Jan 1;8(1):80-105. doi: 10.1504/IJPD.2009.02375
[2]Czerwińska K, Dwornicka R, Pacana A. Improvement of the surface of the combustion chamber of a piston using selected techniques of production organization. Materials Research Proceedings. 2020 Nov 15;17:270-5. doi: 10.21741/9781644901038-40
[3] Qu Z, Shen J, Li R, Liu J, Guan Q. Partsnet: A unified deep network for automotive engine precision parts defect detection. InProceedings of the 2018 2nd International Conference on Computer Science and Artificial Intelligence 2018 Dec 8 (pp. 594-599). doi: 10.48550/arXiv.1810.12061
[4] Islam MR, Zamil MZ, Rayed ME, Kabir MM, Mridha MF, Nishimura S, Shin J. Deep Learning and Computer Vision Techniques for Enhanced Quality Control in Manufacturing Processes. IEEE Access. 2024 Sep 2. doi: 10.1109/ACCESS.2024.3453664
[5] Abagiu MM, Cojocaru D, Manta F, Mariniuc A. Detecting machining defects inside engine piston chamber with computer vision and machine learning. Sensors. 2023 Jan 10;23(2):785. doi: 10.3390/s23020785
[6] Zheng B, Li Y, Zhang J, Wen C, Zhu S, Xu D. Detection of Piston Surface Defects Based on Machine Vision. In2022 IEEE 6th Information Technology and Mechatronics Engineering Conference (ITOEC) 2022 Mar 4 (Vol. 6, pp. 852-857). IEEE. doi: 10.1109/ITOEC53115.2022.9734579
[7] Lin J, Wang D, Tian H, Liu Z. Surface defect detection of machined parts based on machining texture direction. Measurement Science and Technology. 2020 Dec 2;32(2):025204. doi: 10.1088/1361-6501/abb485
[8] Saligheh, A., Hajialimohammadi, A., Abedini, V. Cutting Forces and Tool Wear Investigation for Face Milling of Bimetallic Composite Parts Made of Aluminum and Cast Iron Alloys. International Journal of Engineering, 2020; 33(6): 1142-1148. doi: 10.5829/ije.2020.33.06c.12
[9] Mercy KG, Rao SK. A framework for rail surface defect prediction using machine learning algorithms. In2018 International Conference on Inventive Research in Computing Applications (ICIRCA) 2018 Jul 11 (pp. 972-977). IEEE. doi: 10.1109/ICIRCA.2018.8597394
[10] Židek K, Hošovský A, Dubják J. Diagnostics of surface errors by embedded vision system and its classification by machine learning algorithms. Key Engineering Materials. 2016 Feb 8;669:459-66. doi: 10.4028/www.scientific.net/KEM.669.459
[11] Bhatt PM, Malhan RK, Rajendran P, Shah BC, Thakar S, Yoon YJ, Gupta SK. Image-based surface defect detection using deep learning: A review. Journal of Computing and Information Science in Engineering. 2021 Aug 1;21(4):040801. doi: 10.1115/1.4049535
[12]Hosseini AR, Moosavian A, Javan S, B Shokouhi S. Development of Machine Vision System to Track Movement of an Engine Timing Belt Tensioner Based on Deep Neural Network. The Journal of Engine Research. 2022 Jun 22;67(67):16-23. doi: 10.22034/er.2022.697916
[13]Taghandiki K, Dallakehnejad M, Rahimi Asiabaraki H. Reducing Air Pressure System Repair Costs in Scania Trucks through Deep Learning. Journal of Engineering and Applied Research. 2024 May 7;1(1):183-96.doi: 10.48301/jear.2024.447671.1022
[14]Rahmatinejad B, Rahimi Asiabaraki H, Azimpour Shishevan F. Diagnosing Dimensional Defects and Valve Cracks Using Machine Vision and Acoustic Emission. Karafan Journal. 2023 Nov 22;20(3):149-68.doi: 10.48301/kssa.2023.391572.2501
[15] Siddique N, Paheding S, Elkin CP, Devabhaktuni V. U-net and its variants for medical image segmentation: A review of theory and applications. IEEE access. 2021 Jun 3;9:82031-57. doi: 10.1109/ACCESS.2021.3086020
[16] Rezatofighi H, Tsoi N, Gwak J, Sadeghian A, Reid I, Savarese S. Generalized intersection over union: A metric and a loss for bounding box regression. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition 2019 (pp. 658-666). doi: 10.1109/CVPR.2019.00075
Ranjbar,A. , Hajalimohammadi,A. , Moosavian,A. , Barin,Z. and Rahmani,M. R. (2025). Defect detection in piston rings using deep learning-based computer vision. The Journal of Engine Research, 72(2), 9-18. doi: 10.22034/er.2025.2056649.1080
MLA
Ranjbar,A. , , Hajalimohammadi,A. , , Moosavian,A. , , Barin,Z. , and Rahmani,M. R. . "Defect detection in piston rings using deep learning-based computer vision", The Journal of Engine Research, 72, 2, 2025, 9-18. doi: 10.22034/er.2025.2056649.1080
HARVARD
Ranjbar A., Hajalimohammadi A., Moosavian A., Barin Z., Rahmani M. R. (2025). 'Defect detection in piston rings using deep learning-based computer vision', The Journal of Engine Research, 72(2), pp. 9-18. doi: 10.22034/er.2025.2056649.1080
CHICAGO
A. Ranjbar, A. Hajalimohammadi, A. Moosavian, Z. Barin and M. R. Rahmani, "Defect detection in piston rings using deep learning-based computer vision," The Journal of Engine Research, 72 2 (2025): 9-18, doi: 10.22034/er.2025.2056649.1080
VANCOUVER
Ranjbar A., Hajalimohammadi A., Moosavian A., Barin Z., Rahmani M. R. Defect detection in piston rings using deep learning-based computer vision. Engine, 2025; 72(2): 9-18. doi: 10.22034/er.2025.2056649.1080
