A Lightweight YOLOv11-Based Model for Steel Surface Defect Detection.

Saved in:

Bibliographic Details
Title:	A Lightweight YOLOv11-Based Model for Steel Surface Defect Detection.
Authors:	Wu, Fanghan¹ 1514705165@qq.com, Zhang, Yujun² 1997zyj@163.com
Source:	Engineering Letters. Jul2026, Vol. 34 Issue 7, p2728-2739. 12p.
Subjects:	Industrial applications, Object recognition (Computer vision), Feature extraction, Real-time computing
Abstract:	Steel surface defect detection plays a critical role in ensuring the quality and reliability of steel production and requires both high accuracy and real-time performance. In this study, a lightweight defect detection model based on YOLOv11n is proposed. First, C3K2-PConv is introduced to reduce redundant computation while maintaining effective feature representation. Second, SAConv-Neck is incorporated to enhance multi-scale feature fusion and improve memory efficiency during inference. Third, MBConv-Head is employed to improve prediction efficiency and reduce parameter redundancy. Experimental results on the NEU-DET dataset demonstrate that the proposed model significantly reduces model complexity, with parameters decreasing from 2.59M to 2.04M and computational cost reduced from 3.22 GFLOPs to 2.34 GFLOPs. Meanwhile, the model achieves 77.1% mAP@50 and 192 FPS, outperforming the baseline YOLOv11n. These results demonstrate that the proposed method achieves a favorable balance between detection accuracy and computational efficiency, making it suitable for real-time steel surface defect detection in industrial applications. [ABSTRACT FROM AUTHOR]
	Copyright of Engineering Letters is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Steel surface defect detection plays a critical role in ensuring the quality and reliability of steel production and requires both high accuracy and real-time performance. In this study, a lightweight defect detection model based on YOLOv11n is proposed. First, C3K2-PConv is introduced to reduce redundant computation while maintaining effective feature representation. Second, SAConv-Neck is incorporated to enhance multi-scale feature fusion and improve memory efficiency during inference. Third, MBConv-Head is employed to improve prediction efficiency and reduce parameter redundancy. Experimental results on the NEU-DET dataset demonstrate that the proposed model significantly reduces model complexity, with parameters decreasing from 2.59M to 2.04M and computational cost reduced from 3.22 GFLOPs to 2.34 GFLOPs. Meanwhile, the model achieves 77.1% mAP@50 and 192 FPS, outperforming the baseline YOLOv11n. These results demonstrate that the proposed method achieves a favorable balance between detection accuracy and computational efficiency, making it suitable for real-time steel surface defect detection in industrial applications. [ABSTRACT FROM AUTHOR]
ISSN:	1816093X