DGYOLOv8: An Enhanced Model for Steel Surface Defect Detection Based on YOLOv8

Guanlin Zhu, Honggang Qi () and Ke Lv
Additional contact information
Guanlin Zhu: School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 101408, China
Honggang Qi: School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 101408, China
Ke Lv: School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 101408, China

Mathematics, 2025, vol. 13, issue 5, 1-16

Abstract: The application of deep learning-based defect detection models significantly reduces the workload of workers and enhances the efficiency of inspections. In this paper, an enhanced YOLOv8 model (DCNv4_C2f + GAM + InnerMPDIoU + YOLOv8, hereafter referred to as DGYOLOv8) is developed to tackle the challenges of object detection in steel surface defect detection tasks. DGYOLOv8 incorporates a deformable convolution C2f (DCNv4_C2f) module into the backbone network to allow adaptive adjustment of the receptive field. Additionally, it integrates a Gate Attention Module (GAM) within the spatial and channel attention mechanisms, enhancing feature selection through a gating mechanism that strengthens key features, thereby improving the model’s generalization and interpretability. The InnerMPDIoU, which incorporates the latest Inner concepts, enhances detection accuracy and the ability to handle detailed aspects effectively. This model helps to address the limitations of current networks. Experimental results show improvements in precision (P), recall (R), and mean average precision (mAP) compared to existing models.

Keywords: DGYOLOv8; defect detection; deformable convolutional networks; global attention mechanism; InnerMPDIoU (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/13/5/831/pdf (application/pdf)
https://www.mdpi.com/2227-7390/13/5/831/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:13:y:2025:i:5:p:831-:d:1603762

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().