Steel Surface Defect Detection Method based on YOLOv11-MobileNetv4
DOI:
https://doi.org/10.6919/ICJE.202502_11(2).0002Keywords:
Steel Defect Detection; Deep Learning; YOLO Algorithm; MobileNet; Lightweight Model; Computer Vision.Abstract
Steel occupies a central position in modern industry, but due to long-term use, cracks, corrosion and other defects often appear on its surface, threatening structural safety. Traditional manual inspection methods are inefficient and prone to misdetection and leakage, therefore, efficient and accurate inspection techniques are urgently needed. In this paper, we propose a lightweight steel surface defect detection model YOLOv11-MobilNetv4 based on the combination of YOLOv11 and MobileNetv4. Experimental results show that YOLOv11-MobileNetv4 shows a faster detection speed suitable for application in mobile devices, and the mAP reaches 0.714, which is comparable. This study provides an effective solution for steel defect detection and lays the foundation for subsequent lightweight applications of deep learning models.
Downloads
References
[1] Feng Yingbin, Liu Wenze. Steel Surface Defect Detection Algorithm Based on HPDE-YOLO [J]. Journal of Shenyang Ligong University, 2025, 44(1): 31-38. (In Chinese)
[2] LIU Wenzhao, ZHANG Dan. Lightweight steel surface defect detection model based on machine vision[J/OL]. Computing Technology and Automation, 2024, 43(3): 43-49. DOI:10.16339/j.cnki. jsjsyzdh. 202403008. (In Chinese)
[3] Tao Yufeng, Jiang Lin, Da Mei. Algorithm for steel surface defect detection based on RCD-YOLO[J/OL]. Control Engineering of China: 1-9. DOI:10.14107/j.cnki.kzgc.20240426. (In Chinese)
[4] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-Based Learning Applied to Document Recognition[J/OL]. Proceedings of the IEEE, 1998, 86: 2278-2324. DOI:10.1109/5.726791.
[5] LECUN Y, BENGIO Y, HINTON G. Deep learning [J/OL]. Nature, 2015, 521(7553): 436-444. doi:10.1038/nature14539.
[6] HE Y, SONG K, MENG Q, et al. An End-to-End Steel Surface Defect Detection Approach via Fusing Multiple Hierarchical Features[J/OL]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(4): 1493-1504. doi:10.1109/TIM.2019.2915404.
[7] ULTRALYTICS. yolo11 ???? new [EB/OL]. [2025-01-10]. https://docs.ultralytics.com/zh/models/yolo11.
[8] QIN D, LEICHNER C, DELAKIS M, et al. MobileNetV4 - Universal Models for the Mobile Ecosystem [A/OL]. arXiv, 2024[2025-01-10]. http://arxiv.org/abs/2404.10518. DOI:10.48550/arXiv.2404.10518.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
