Rail Surface Defect Detection with Multi-Scale Detection Heads and Receptive Field Enhancement
Academic Journal of Emerging Technologies · 18.11.2026 · 4 (1) · 76
DOI: 10.63313/ajet.9081 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academic Journal of Emerging Technologies, online erschienen am 18.11.2026. Verfasst von Yuhao Ma und 4 weiteren Autor:innen. Die Arbeit nennt 6 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Rail surface defects have diverse types and large scale variations, and standard detectors often miss fine-grained texture defects such as corrugation. This paper proposes a lightweight rail defect detection model based on an improved YOLOv11n. Three progressive improvements are introduced. First, a high-resolution P2 detection head is added to the neck to build a four-scale detection architecture, enhancing the perception of fine-grained textures. Second, C3k2 modules in the deep backbone are replaced with C2f modules, which refine gradient flow and deep semantic feature extraction while alleviating false detections caused by the shallow head. Third, an SPPF module is introduced in the P2 branch to enlarge the receptive field and compensate for weak semantics in shallow features. On a self-built Rail Surface Defect Dataset, the final model achieves 0.797 mAP@0.5 and 0.562 mAP@0.5-0.95 on the independent test set, improving the baseline by 11.78% and 14.93%, respectively. For corrugation, which is severely missed by the baseline, accuracy increases from 0.292 to 0.581. The model size increases only from 5.51 MB to 6.07 MB, achieving a good balance between accuracy and deployment cost for high-frequency, large-scale automated railway inspection.
Zitieren
APA
Ma, Y., Li, J., Hu, Z., Wang, T., & Xu, J. (2026). Rail Surface Defect Detection with Multi-Scale Detection Heads and Receptive Field Enhancement. Academic Journal of Emerging Technologies, 4(1), 76. https://doi.org/10.63313/ajet.9081
Vancouver
Ma Y, Li J, Hu Z, Wang T, Xu J. Rail Surface Defect Detection with Multi-Scale Detection Heads and Receptive Field Enhancement. Academic Journal of Emerging Technologies. 2026;4(1):76. doi:10.63313/ajet.9081
BibTeX
@article{Ma2026,
author = {Ma, Yuhao and Li, Jincheng and Hu, Zhichao and Wang, Taoyu and Xu, Jie},
title = {Rail Surface Defect Detection with Multi-Scale Detection Heads and Receptive Field Enhancement},
journal = {Academic Journal of Emerging Technologies},
year = {2026},
volume = {4},
number = {1},
pages = {76},
doi = {10.63313/ajet.9081}
}
EUVIMED