Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Computing, Performance and Communication Systems
  4. Vol 10, Issue 1, 2026

ECD-YOLO: Multiscale Feature Analysis and Detail-Preserving Downsampling for Steel Surface Defect Detection

Download as PDF

DOI: 10.23977/cpcs.2026.100109 | Downloads: 0 | Views: 15

Author(s)

Zhang Xiaojian 1

Affiliation(s)

1 School of Management, University of Shanghai for Science and Technology, Shanghai, 200093, China

Corresponding Author

Zhang Xiaojian

ABSTRACT

Automated inspection of steel surfaces is challenged by minute defects, weak contrast, repetitive rolling textures, and substantial scale variation. This paper presents ECD-YOLO, an enhanced YOLO11n detector for accurate and efficient defect localization. The proposed ETC-Down module decomposes downsampling into four complementary branches that preserve spatial structure, gradient edges, multiscale texture, and local contrast. The DAFB module employs differential attention to suppress redundant background responses and combines large-kernel positional encoding, dilated depthwise convolution, and channel gating to strengthen global-local discrimination. SSFF and TFE are incorporated into the neck to align and fuse P3-P5 features across spatial scales. Experiments on the NEU-DET dataset show that, relative to YOLO11n, ECD-YOLO improves precision, recall, [email protected], and [email protected]:0.95 by 12.1, 2.6, 5.4, and 10.2 percentage points, respectively, while increasing the parameter count from 2.58 M to 3.28 M and computation from 6.3 to 7.4 GFLOPs. Ablation results confirm complementary gains in detail preservation, background suppression, and multiscale fusion. ECD-YOLO therefore provides a favorable accuracy-efficiency trade-off for online steel surface inspection.

KEYWORDS

Steel surface defect detection; YOLO11; Differential attention; Multiscale feature fusion; Detail-preserving downsampling

CITE THIS PAPER

Zhang Xiaojian. ECD-YOLO: Multiscale Feature Analysis and Detail-Preserving Downsampling for Steel Surface Defect Detection. Computing, Performance and Communication Systems (2026). Vol. 10, No. 1, 76-86. DOI: http://dx.doi.org/10.23977/cpcs.2026.100109.

REFERENCES

[1] D. H. Wu, D. H. You, Z. L. Liu, et al., "Mechanism and experimental research on skin depth in AC magnetic flux leakage testing," Chinese Journal of Scientific Instrument, vol. 35, no. 2, pp. 327-336, 2014.
[2] X. Y. Wei, W. Y. Liu, L. Y. Wang, et al., "Surface inspection of polished rail welds based on photometric stereo," Journal of Mechanical Engineering, vol. 60, no. 24, pp. 11-24, 2024.
[3] S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards real-time object detection with region proposal networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 2017.
[4] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: Unified, real-time object detection," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 2016, pp. 779-788.
[5] F. Q. Lei, L. W. Ma, P. Guan, et al., "Research on surface defect detection algorithm of steel equipment based on DFD-YOLOv11n," Computer Engineering and Applications, vol. 62, no. 2, pp. 92-102, 2026.
[6] J. J. Ma, J. H. Zhang, Q. Wang, et al., "Strip surface defect detection algorithm based on improved YOLOv11," Journal of Iron and Steel Research, vol. 37, no. 10, pp. 1345-1358, 2025.
[7] Z. Zeng, X. L. Shi, H. Yan, et al., "RGCE-YOLO: Lightweight steel surface defect detection algorithm," Opto-Electronic Engineering, vol. 53, no. 3, Art. no. 250240, 2026.
[8] S. Gao, M. Chu, and L. Zhang, "A detection network for small defects of steel surface based on YOLOv7," Digital Signal Processing, vol. 149, Art. no. 104484, 2024.
[9] G. Liu, M. Chu, R. Gong, et al., "Global attention module and cascade fusion network for steel surface defect detection," Pattern Recognition, vol. 158, Art. no. 110979, 2025.
[10] T. H. Liu and Z. S. He, "Surface defect detection method based on auto-encoding and knowledge distillation," Journal of Computer Applications, vol. 41, no. 11, pp. 3200-3205, 2021.
[11] Z. P. Ji, C. W. Lin, Z. W. He, et al., "Strip steel surface defect detection integrating global-local awareness and hierarchical feature fusion," Computer Engineering and Applications, vol. 62, no. 4, pp. 373-381, 2026, doi: 10.3778/j.issn.1002-8331.2412-0407.
[12] G. Jocher and J. Qiu, "Ultralytics YOLO11, version 11.0.0," computer program, 2024.
[13] G. Jocher, "Ultralytics YOLOv5: Version 7.0," Zenodo, 2022.
[14] M. Kang, C. M. Ting, and F. F. Ting, "ASF-YOLO: A novel YOLO model with attentional scale sequence fusion for cell instance segmentation," Image and Vision Computing, vol. 147, Art. no. 105057, 2024.
[15] B. Zhang and R. Sennrich, "Root mean square layer normalization," in Advances in Neural Information Processing Systems, vol. 32, 2019, pp. 12360-12371.
[16] C. Li, L. Li, H. Jiang, et al., "YOLOv6: A single-stage object detection framework for industrial applications," arXiv:2209.02976, 2022.
[17] G. Jocher, A. Chaurasia, and J. Qiu, "Ultralytics YOLOv8: Version 8.0.0," computer program, 2023.
[18] C. Y. Wang, I. H. Yeh, and H. Y. M. Liao, "YOLOv9: Learning what you want to learn using programmable gradient information," in Computer Vision-ECCV 2024, Cham, Switzerland: Springer, 2024, pp. 1-21.
[19] Wang, H. Chen, L. Liu, et al., "YOLOv10: Real-time end-to-end object detection," in Advances in Neural Information Processing Systems, vol. 37, 2024, pp. 107984-108011.
[20] Y. Tian, Q. Ye, and D. Doermann, "YOLOv12: Attention-centric real-time object detectors," in Advances in Neural Information Processing Systems, vol. 38, 2025, pp. 78433-78451.
[21] Y. Zhao, W. Lv, S. Xu, et al., "DETRs beat YOLOs on real-time object detection," in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition, Seattle, WA, USA, 2024, pp. 16965-16974.
[22] T. Ye, L. Dong, Y. Xia, et al., "Differential transformer," in Proc. International Conference on Learning Representations, 2025.
[23] R. Sunkara and T. Luo, "No more strided convolutions or pooling: A new CNN building block for low-resolution images and small objects," in Machine Learning and Knowledge Discovery in Databases, Cham, Switzerland: Springer, 2023, pp. 443-459.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 4189
Visits: 301682

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.