Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection

Ethem Belka Şahin; Seçkin Arı

doi:10.35377/saucis...1945069

Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection

Abstract

Metal surface defect detection is a fundamental quality control challenge in the manufacturing industry. Surface defects, such as holes and crescent-shaped edge defects, directly affect the structural integrity of metal components, making their accurate and timely identification critical in the production pipeline. Single-model deep learning approaches tend to exhibit inconsistent performance across varying imaging conditions, such as blur, noise, and illumination changes, motivating the investigation of multi-model fusion strategies. This paper proposes a hybrid object detection system that integrates YOLOv8s and Faster R-CNN through a Sugeno-type fuzzy inference mechanism for metal surface defect detection. The system operates in three sequential stages: independent inference by both models, fuzzy-weighted fusion using a rule-based confidence evaluation mechanism, and hybrid non-maximum suppression. A greedy intersection-over-union matching algorithm pairs the predictions from both models, and the resulting confidence scores are fuzzified through a Sugeno-type fuzzy inference system to produce adaptive fusion weights. A parametric ablation study was conducted across nine system configurations by varying the membership function boundaries, unmatched detection filters, class-specific rule tables, and hybrid NMS thresholds. The system was evaluated on a custom metal surface dataset containing two defect classes under seven types of synthetic image distortion across 26 sub-levels comprising 2,080 test images. The best-performing configuration achieved a macro-average F1 score of 0.9100 and precision of 0.9437, outperforming four comparison methods, including the standard NMS ensemble, equal-weight fusion, Maximum Confidence Selection, and Weighted Box Fusion. The hybrid system demonstrated superior robustness to Gaussian blur and motion blur conditions compared to either standalone model, with an inference latency of 87.9 ms at 11.4 FPS, confirming its suitability for production-line deployment.

Keywords

Supporting Institution

Sakarya University

Ethical Statement

It is declared that during the preparation process of this study, scientific and ethical principles were followed and all studies benefited from are stated in the bibliography.

Thanks

The authors thank Sakarya University for providing the computational infrastructure used in this study.

References

R. Usamentiaga, D. G. Lema, O. D. Pedrayes, and D. F. Garcia, “Automated Surface Defect Detection in Metals: A Comparative Review of Object Detection and Semantic Segmentation Using Deep Learning,” IEEE Trans. Ind. Appl., vol. 58, no. 3, pp. 4203–4213, 2022, doi: 10.1109/TIA.2022.3151560.
X. Wen, J. Shan, Y. He, and K. Song, “Steel Surface Defect Recognition: A Survey,” Coatings, vol. 13, no. 1, p. 17, 2022, doi: 10.3390/COATINGS13010017.
X. Zheng, S. Zheng, Y. Kong, and J. Chen, “Recent advances in surface defect inspection of industrial products using deep learning techniques,” Int. J. Adv. Manuf. Technol., vol. 113, no. 1–2, pp. 35–58, 2021, doi: 10.1007/S00170-021-06592-8.
Y. Chen, Y. Ding, F. Zhao, E. Zhang, Z. Wu, and L. Shao, “Surface Defect Detection Methods for Industrial Products: A Review,” Appl. Sci., vol. 11, no. 16, p. 7657, 2021, doi: 10.3390/APP11167657.
E. Cumbajin et al., “A Systematic Review on Deep Learning with CNNs Applied to Surface Defect Detection,” J. Imaging, vol. 9, no. 10, p. 193, 2023, doi: 10.3390/JIMAGING9100193.
J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), 2016, pp. 779–788.
A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” arXiv preprint arXiv:2004.10934, 2020.
S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 6, pp. 1137–1149, 2017, doi: 10.1109/TPAMI.2016.2577031.

R. Girshick, “Fast R-CNN,” in Proc. IEEE Int. Conf. Comput. Vis. (ICCV), 2015, pp. 1440–1448, doi: 10.1109/ICCV.2015.169.
Z. Li, X. Wei, M. Hassaballah, Y. Li, and X. Jiang, “A deep learning model for steel surface defect detection,” Complex Intell. Syst., vol. 10, no. 1, pp. 885–897, 2024, doi: 10.1007/S40747-023-01180-7.
W. Zhao, F. Chen, H. Huang, D. Li, and W. Cheng, “A New Steel Defect Detection Algorithm Based on Deep Learning,” Comput. Intell. Neurosci., vol. 2021, p. 5592878, 2021, doi: 10.1155/2021/5592878.
D. Yang, Y. Cui, Z. Yu, and H. Yuan, “Deep Learning Based Steel Pipe Weld Defect Detection,” Appl. Artif. Intell., vol. 35, no. 15, pp. 1237–1249, 2021, doi: 10.1080/08839514.2021.1975391.
T. T. H. Vu, D. L. Pham, and T.-W. Chang, “A YOLO-based Real-time Packaging Defect Detection System,” Procedia Comput. Sci., vol. 217, pp. 886–894, 2023, doi: 10.1016/J.PROCS.2022.12.285.
T. G. Dietterich, “Ensemble Methods in Machine Learning,” in Multiple Classifier Systems, vol. 1857, Springer, 2000, pp. 1–15, doi: 10.1007/3-540-45014-9_1.
Z.-H. Zhou, Ensemble Methods: Foundations and Algorithms. Chapman and Hall/CRC, 2012, doi: 10.1201/b12207.
A. Neubeck and L. Van Gool, “Efficient Non-Maximum Suppression,” in Proc. 18th Int. Conf. Pattern Recognit. (ICPR), vol. 3, 2006, pp. 850–855, doi: 10.1109/ICPR.2006.479.
R. Solovyev, W. Wang, and T. Gabruseva, “Weighted Boxes Fusion: Ensembling Boxes from Different Object Detection Models,” Image Vis. Comput., vol. 107, p. 104117, 2021, doi: 10.1016/j.imavis.2021.104117.
Y. Zheng, Z. Xu, and X. Wang, “The Fusion of Deep Learning and Fuzzy Systems: A State-of-the-Art Survey,” IEEE Trans. Fuzzy Syst., vol. 30, no. 8, pp. 2783–2799, 2022, doi: 10.1109/TFUZZ.2021.3062899.
W. Ding et al., “FTransCNN: Fusing Transformer and a CNN Based on Fuzzy Logic for Uncertain Medical Image Segmentation,” Inf. Fusion, vol. 99, p. 101880, 2023, doi: 10.1016/J.INFFUS.2023.101880.
A. Halder et al., “A fuzzy rank-based deep ensemble methodology for multi-class skin cancer classification,” Sci. Rep., vol. 15, no. 1, 2025, doi: 10.1038/s41598-025-90423-3.
A. Dheeraj, R. K. Chaturvedi, S. Nigam, M. A. Haque, and S. Marwaha, “Fuzzy rank fusion of deep neural networks for weed identification in groundnut crop,” Discov. Sustain., vol. 6, no. 1, 2025, doi: 10.1007/s43621-025-01705-9.
L. A. Zadeh, “Fuzzy Sets,” Inf. Control, vol. 8, no. 3, pp. 338–353, 1965, doi: 10.1016/S0019-9958(65)90241-X.
T. Takagi and M. Sugeno, “Fuzzy Identification of Systems and Its Applications to Modeling and Control,” IEEE Trans. Syst. Man Cybern., vol. 15, no. 1, pp. 116–132, 1985, doi: 10.1109/TSMC.1985.6313399.
Ultralytics, “ultralytics/ultralytics: Ultralytics YOLO — GitHub Repository,” 2026. [Online]. Available: https://github.com/ultralytics/ultralytics

Details

Primary Language

English

Subjects

Artificial Intelligence (Other)

Journal Section

Research Article

Authors

Ethem Belka Şahin ^*
0009-0008-0234-6941
Türkiye

Seçkin Arı
0000-0001-7556-8205
Türkiye

Early Pub Date

June 22, 2026

Publication Date

June 30, 2026

Submission Date

May 5, 2026

Acceptance Date

May 18, 2026

Published in Issue

Year 2026 Volume: 9 Number: 3

DOI

https://doi.org/10.35377/saucis...1945069

IZ

https://izlik.org/JA36WM24SX

Cite

RIS / Bibtex

APA

Şahin, E. B., & Arı, S. (2026). Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection. Sakarya University Journal of Computer and Information Sciences, 9(3), 829-839. https://doi.org/10.35377/saucis...1945069

AMA

1.Şahin EB, Arı S. Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection. SAUCIS. 2026;9(3):829-839. doi:10.35377/saucis.1945069

Chicago

Şahin, Ethem Belka, and Seçkin Arı. 2026. “Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection”. Sakarya University Journal of Computer and Information Sciences 9 (3): 829-39. https://doi.org/10.35377/saucis. 1945069.

EndNote

Şahin EB, Arı S (June 1, 2026) Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection. Sakarya University Journal of Computer and Information Sciences 9 3 829–839.

IEEE

[1]E. B. Şahin and S. Arı, “Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection”, SAUCIS, vol. 9, no. 3, pp. 829–839, June 2026, doi: 10.35377/saucis...1945069.

ISNAD

Şahin, Ethem Belka - Arı, Seçkin. “Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection”. Sakarya University Journal of Computer and Information Sciences 9/3 (June 1, 2026): 829-839. https://doi.org/10.35377/saucis. 1945069.

JAMA

1.Şahin EB, Arı S. Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection. SAUCIS. 2026;9:829–839.

MLA

Şahin, Ethem Belka, and Seçkin Arı. “Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection”. Sakarya University Journal of Computer and Information Sciences, vol. 9, no. 3, June 2026, pp. 829-3, doi:10.35377/saucis. 1945069.

Vancouver

1.Ethem Belka Şahin, Seçkin Arı. Fuzzy Logic-Based Hybrid Deep Learning System for Metal Surface Defect Detection. SAUCIS. 2026 Jun. 1;9(3):829-3. doi:10.35377/saucis. 1945069