Research on Rail Surface Defect Detection Based on Improved CenterNet
Electronics · 2024-09-09 · 13 (17) · 3580
DOI: 10.3390/electronics13173580 ↗At a glance
Journal article in Electronics, published online on 2024-09-09. Written by Yizhou Mao and 4 co-authors. Cited 11 times so far (Crossref); 35 references. The full text is freely available (open access, CC BY 4.0). Funded by: National Natural Science Foundation of China.
Abstract (original language)
Rail surface defect detection is vital for railway safety. Traditional methods falter with varying defect sizes and complex backgrounds, while two-stage deep learning models, though accurate, lack real-time capabilities. To overcome these challenges, we propose an enhanced one-stage detection model based on CenterNet. We replace ResNet with ResNeXt and implement a multi-branch structure for better low-level feature extraction. Additionally, we integrate SKNet attention mechanism with the C2f structure from YOLOv8, improving the model’s focus on critical image regions and enhancing the detection of minor defects. We also introduce an elliptical Gaussian kernel for size regression loss, better representing the aspect ratio of rail defects. This approach enhances detection accuracy and speeds up training. Our model achieves a mean accuracy (mAP) of 0.952 on the rail defects dataset, outperforming other models with a 6.6% improvement over the original and a 35.5% increase in training speed. These results demonstrate the efficiency and reliability of our method for rail defect detection.
Funding
- National Natural Science Foundation of China
Cite
APA
Mao, Y., Zheng, S., Li, L., Shi, R., & An, X. (2024). Research on Rail Surface Defect Detection Based on Improved CenterNet. Electronics, 13(17), 3580. https://doi.org/10.3390/electronics13173580
Vancouver
Mao Y, Zheng S, Li L, Shi R, An X. Research on Rail Surface Defect Detection Based on Improved CenterNet. Electronics. 2024;13(17):3580. doi:10.3390/electronics13173580
BibTeX
@article{Mao2024,
author = {Mao, Yizhou and Zheng, Shubin and Li, Liming and Shi, Renjie and An, Xiaoxue},
title = {Research on Rail Surface Defect Detection Based on Improved CenterNet},
journal = {Electronics},
year = {2024},
volume = {13},
number = {17},
pages = {3580},
doi = {10.3390/electronics13173580}
}
EUVIMED