OS9-1 A Study on Surface Defect Detection Algorithm of Strip Steel Based on YOLOv8n

Title:	OS9-1 A Study on Surface Defect Detection Algorithm of Strip Steel Based on YOLOv8n
Publication:	ICAROB2025
Volume:	30
Pages:	241-247
ISSN:	2188-7829
DOI:	10.5954/ICAROB.2025.OS9-1
Author(s):	Haozhe Sun, Fengzhi Dai, Junjin Chen
Publication Date:	February 13, 2025
Keywords:	Deep learning, Surface defect detection, YOLOv8n, Dynamic Snake Convolution, Efficient Multi-Scale Attention Module
Abstract:	Hot rolled steel strip has been extensively applied in industrial production and processing due to its outstanding properties. Nevertheless, during the production procedure, as a result of technological constraints, defects will inevitably occur on the surface of the steel strip, significantly influencing the performance and safety of the steel strip. Hence, how to detect the surface defects of steel strips has turned into the key point. In this paper, an enhanced YOLOv8n network model is proposed to make it applicable for the surface defect detection tasks of hot rolled steel strips. The improved model introduces Dynamic Snake Convolution and Efficient Multi-Scale Attention Module. The average precision of the improved model is 6.8 percentage points higher than that of the original model.
PDF File:	https://alife-robotics.co.jp/members2025/icarob/data/html/data/OS/OS9/OS9-1.pdf
Copyright:	© The authors. This article is distributed under the terms of the Creative Commons Attribution License 4.0, which permits non-commercial use, distribution and reproduction in any medium, provided the original work is properly cited. See for details: https://creativecommons.org/licenses/by-nc/4.0/

HOME