A novel steel surface defect detector with wavelet upsampling and frequency-domain attention
Source abstract
Abstract
Steel surface defect detection is a crucial step in improving product quality, ensuring safety, and enhancing production efficiency. However, existing object detection models still face challenges such as high miss rates and computational complexity when handling diverse defect types and complex background noise. To address these issues, this study proposes a steel surface defect detection model based on the real-time detection transformer, aimed at improving detection accuracy in complex industrial environments. To tackle the challenge of significant shape variations among defects, a dynamic scale aggregation module is designed to fuse critical information from multiple pathways, enabling efficient multi-scale feature extraction. Additionally, a frequency-enhanced dynamic attention mechanism is proposed, utilising the fast Fourier transform for global context modelling in the frequency domain, enhancing the joint perception of low-frequency structural and high-frequency detailed features. Finally, a wavelet upsampling operator is developed to decompose input features into high- and low-frequency components, enabling targeted refinement and adaptive fusion. Experimental results show that the improved model achieves AP of 46.6% and 36.0% on the NEU-DET and GC10-DET datasets, respectively, surpassing the baseline by 2.7% and 1.9%, while reducing FLOPs by 10.0 G. This contributes to more reliable surface quality control in steel processing workflows.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Surface Roughness and Optical Measurements · Advanced Neural Network Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Hubei Normal University
Funding
National Natural Science Foundation of China
References (76)
Show source-provided references
1. 10.1007/s12239-021-0102-710.1007/s12239-021-0102-7 ↗
2. 10.1007/s10845-020-01733-410.1007/s10845-020-01733-4 ↗
3. 10.1016/j.measurement.2023.11277610.1016/j.measurement.2023.112776 ↗
4. 10.1088/1361-6501/ace92910.1088/1361-6501/ace929 ↗
5. 10.1080/10589759.2024.240844110.1080/10589759.2024.2408441 ↗
6. 10.1007/s10921-023-00977-310.1007/s10921-023-00977-3 ↗
7. 10.3390/app1412517110.3390/app14125171 ↗
8. 10.1784/insi.2023.65.1.1910.1784/insi.2023.65.1.19 ↗
9. 10.1007/s00170-021-08543-910.1007/s00170-021-08543-9 ↗
10. 10.3390/met1203043210.3390/met12030432 ↗
11. 10.1007/s11668-024-02046-x10.1007/s11668-024-02046-x ↗
12. Song F Bao K Deng M et al. Steel surface defect detection based on PSO-Gabor and an improved faster RCNN. Nondestructive Testing and Evaluation. 2025. p. 1–24.10.1080/10589759.2025.2512553 ↗
13. 10.1080/10589759.2025.247410310.1080/10589759.2025.2474103 ↗
14. 10.1080/09544828.2025.250430410.1080/09544828.2025.2504304 ↗
15. 10.1016/j.jmsy.2025.06.01410.1016/j.jmsy.2025.06.014 ↗
16. 10.1080/10589759.2024.239320510.1080/10589759.2024.2393205 ↗
17. 10.1109/TPAMI.2016.257703110.1109/TPAMI.2016.2577031 ↗
18. Lin T-Y, Goyal P, Girshick R, et al. In focal loss for dense object detection. In: Proceedings of the IEEE international conference on computer vision; Venice, Italy; 22–29 October, 2017. 2017. p. 2980–2988.
19. Li T, Sun Y. In defect detection of solar panels using improved faster R-CNN. In: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), SPIE; Changsha, China. 2023; SPIE Digital Library. p. 605–610.
20. 10.1177/1558925022108664710.1177/15589250221086647 ↗
21. 10.3390/app13201147310.3390/app132011473 ↗
22. 10.1016/j.measurement.2025.11684710.1016/j.measurement.2025.116847 ↗
23. Carion N, Massa F, Synnaeve G, et al. In end-to-end object detection with transformers. In: European conference on computer vision; Glasgow, United Kingdom; 03 November 2020. Springer; 2020. p. 213–229.
24. 10.3390/s2413426210.3390/s24134262 ↗
25. 10.1049/ipr2.1325010.1049/ipr2.13250 ↗
26. 10.1109/ACCESS.2022.322720510.1109/ACCESS.2022.3227205 ↗
27. Feng Y, Huang J, Du S, et al. Hyper-yolo: when visual object detection meets hypergraph computation. In: IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024; IEEE.
28. 10.3390/agriculture1412225610.3390/agriculture14122256 ↗
29. 10.1016/j.microc.2024.11178910.1016/j.microc.2024.111789 ↗
30. Kong L, Dong J, Ge J, et al. In efficient frequency domain-based transformers for high-quality image deblurring. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; Vancouver, BC, Canada. 2023; IEEE. p. 5886–5895.
31. 10.1016/j.ecoinf.2023.10221110.1016/j.ecoinf.2023.102211 ↗
32. 10.3390/electronics1215321010.3390/electronics12153210 ↗
33. Li W, Guo H, Liu X, et al. In efficient face super-resolution via wavelet-based feature enhancement network. In: Proceedings of the 32nd ACM International Conference on Multimedia; Melbourne, Australia. 2024; ACM. p. 4515–4523.10.1145/3664647.3681088 ↗
34. Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. Adv Neural Inf Process Syst. 2017;30.
35. 10.3390/app14221069010.3390/app142210690 ↗
36. Chen J, Kao S-H, He H, et al. In run, don’t walk: chasing higher FLOPS for faster neural networks. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Vancouver, BC, Canada. 2023; IEEE. p. 12021–12031.
37. Lin T-Y, Maire M, Belongie S, et al. In Microsoft COCO: common objects in context, computer vision–ECCV 2014: 13th European Conference. Vol. 13. Zurich, Switzerland: Springer; 2014 p. 740–755. Sep 6–12, 2014.
38. 10.1109/TIM.2019.291540410.1109/TIM.2019.2915404 ↗
39. 10.3390/s2006156210.3390/s20061562 ↗
40. Wang A, Chen H, Lin Z, et al. In Repvit: revisiting mobile CNN from ViT perspective. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; Seattle, WA, USA. 2024; IEEE. p. 15909–15920.
41. Liu Z, Lin Y, Cao Y, et al. In Swin Transformer: hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE/CVF international conference on computer vision. 2021; IEEE. p. 10012–10022.
42. 10.5281/zenodo.636190610.5281/zenodo.6361906 ↗
43. Qin D, Leichner C, Delakis M, et al. Mobilenetv4-universal models for the mobile ecosystem. arXiv. 2024:15098. arXiv:2404.10518.
44. Li Y, Li X, Dai Y, et al. Lsknet: a foundation lightweight backbone for remote sensing: y. Li et al. Int J Comput Vision. 2025;133:1410–1431.10.1007/s11263-024-02247-9 ↗
45. Ding X, Zhang Y, Ge Y, et al. In unireplknet: a universal perception large-kernel convnet for audio video point cloud time-series and image recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; Seattle, WA, USA. 2024; IEEE. p. 5513–5524.
46. Xia Z, Pan X, Song S, et al. In vision transformer with deformable attention. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; New Orleans, Louisiana, USA. 2022; IEEE. p. 4794–4803.
47. Sun S, Ren W, Gao X, et al. In restoring images in adverse weather conditions via histogram transformer. In: Leonardis A, Ricci E, Roth S, Russakovsky O, Sattler T, Varol G, editors. European Conference on Computer Vision; Cham, Switzerland. Springer; 2024. p. 111–129.
48. Guo J Chen X Tang Y et al. Slab: efficient transformers with simplified linear attention and progressive re-parameterized batch normalization. arXiv preprint arXiv:2405.11582. 2024;667:16802–16812.
49. 10.1109/TGRS.2023.331464110.1109/TGRS.2023.3314641 ↗
50. Selvaraju RR, Cogswell M, Das A, et al. In grad-cam: visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE international conference on computer vision; Venice, Italy. 2017; IEEE. p. 618–626.
51. Wang J, Chen K, Xu R, et al. In: Carafe: Content-aware reassembly of features. In: Proceedings of the IEEE/CVF international conference on computer vision; Seoul, Korea (South). 2019. p. 3007–3016.
52. Liu W, Lu H, Fu H, et al. In learning to upsample by learning to sample. In: Proceedings of the IEEE/CVF International Conference on Computer Vision; Paris, France. 2023. p. 6027–6037.
53. Kirkland EJ, Kirkland EJ. Bilinear interpolation. Adv Comput Electron microscopy. 2010:261–263.10.1007/978-1-4419-6533-2_12 ↗
54. Rahman MM, Munir M, Marculescu R. In emcad: efficient multi-scale convolutional attention decoding for medical image segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; Seattle, WA, USA. 2024. p. 11769–11779.
55. Liu W, Anguelov D, Erhan D, et al. In SSD: single shot multibox detector. In: Computer Vision–ECCV 2016: 14th European Conference; Amsterdam, The Netherlands. Springer; 2016 Oct 11–14, 2016, Proceedings, Part I 14. p. 21–37.10.1007/978-3-319-46448-0_2 ↗
56. Li X, Wang W, Wu L, et al. Generalized focal loss: learning qualified and distributed bounding boxes for dense object detection. Adv Neural Inf Process Syst. 2020;33:21002–21012.
57. Zhang S, Chi C, Yao Y, et al. In bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Seattle, WA, USA. 2020. p. 9759–9768.
58. Feng C, Zhong Y, Gao Y, et al. In Tood: Task-aligned one-stage object detection, 2021 IEEE/CVF International Conference on Computer Vision (ICCV); Montreal, QC, Canada. In IEEE Computer Society; 2021. p. 3490–3499.
59. Chen Z, Yang C, Li Q, et al. In disentangle your dense object detector. In: Proceedings of the 29th ACM international conference on multimedia; Chengdu, China. 2021. p. 4939–4948.10.1145/3474085.3475351 ↗
60. Tian Z, Shen C, Chen H, et al. In FCOS: fully convolutional one-stage object detection. In: Proceedings of the IEEE/CVF international conference on computer vision; Seoul, Korea (South). 2019. p. 9627–9636.
61. Zhang H, Wang Y, Dayoub F, et al. In varifocalnet: an IOU-aware dense object detector. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Nashville, TN, USA. 2021. p. 8514–8523.
62. Duan K, Bai S, Xie L, et al. In centernet: keypoint triplets for object detection. In: Proceedings of the IEEE/CVF international conference on computer vision; Seoul, Korea (South). 2019. p. 6569–6578.
63. Lu X, Li B, Yue Y, et al. In: Grid r-cnn. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Long Beach, CA, USA. 2019. p. 7363–7372.
64. Pang J, Chen K, Shi J, et al. In Libra R-CNN: towards balanced learning for object detection. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Long Beach, CA, USA. 2019. p. 821–830.
65. Cai Z, Vasconcelos N. In cascade r-cnn: delving into high quality object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition; Salt Lake City, UT, USA. 2018. p. 6154–6162.
66. Chen Q, Wang Y, Yang T, et al. In: You only look one-level feature, Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Nashville, TN, USA. 2021. p. 13039–13048.
67. Ge Z, Liu S, Wang F, et al. J Yolox: Exceeding yolo series in 2021. arXiv (2021)arXiv:2107.08430.
68. Jocher G, Chaurasia A, Stoken A, et al. Ultralytics/yolov5: v6. 1-tensorrt, tensorflow edge tpu and openvino export and inference. Zenodo. 2022.
69. Li C, Li L, Jiang H, et al. Yolov6: a single-stage object detection framework for industrial applications. arXiv. 2022 arXiv:2209.02976.
70. Wang C-Y, Yeh I-H, Liao H-YM. Yolov9: learning what you want to learn using programmable gradient information. arXiv. 2024;1–21. arXiv:2402.13616.
71. Wang A, Chen H, Liu L, et al. Yolov10: real-time end-to-end object detection. Proceedings of the 38th International Conference on Neural Information Processing Systems; Vancouver, Canada. Vol. 3429. 2024 ; NIPS.
72. Jocher G. In Ultralytics Yolo11. 2024. Available from: https://github.com/ultralytics/ultralytics
73. Liu S Li F Zhang H et al. Dab-detr: dynamic anchor boxes are better queries for detr. arXiv preprint arXiv:2201.12329 2022.
74. Zhang S, Wang X, Wang J, et al. In dense distinct query for end-to-end object detection. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Vancouver, BC, Canada. 2023. p. 7329–7338.
75. Zhang H Li F Liu S et al. Dino: detr with improved denoising anchor boxes for end-to-end object detection. arXiv preprint arXiv:2203.03605 2022.
76. Zhao Y, Lv W, Xu S, et al. In detrs beat yolos on real-time object detection. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; Seattle, WA, USA. 2024. p. 16965–16974.