Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Advances in Computer, Signals and Systems
  4. Vol 10, Issue 2, 2026

Research on Foreign Object Detection Algorithm for High-Speed Railway Catenary Based on Improved YOLOv8n

Download as PDF

DOI: 10.23977/acss.2026.100201 | Downloads: 0 | Views: 22

Author(s)

Jian Sun 1, Zhenhua Wang 1, Tianijng Zhang 1, Siqi Chen 1, Rui Wang 1

Affiliation(s)

1 China Railway Xi'an Group Co., Ltd., Xi'an, 710000, China

Corresponding Author

Jian Sun

ABSTRACT

With the rapid development of industrialization, the demand for railway transportation has been increasing. In this context, the detection of foreign objects in railway catenary systems remains a significant challenge to ensuring the safe operation of railways. To address the issues of low accuracy and poor real-time performance in detecting foreign objects in railway catenary systems, this paper proposes a foreign object detection algorithm based on the AFPN-YOLOv8n deep learning model. To fundamentally improve detection accuracy, the algorithm introduces a feature pyramid network (AFPN) in the Head module, effectively integrating low-level detail features with high-level semantic features, thereby enhancing the network's ability to detect foreign objects in high-speed railway catenary images. Additionally, an efficient multi-scale attention (EMA) module is added after the C3 layer in the Backbone module, further improving the network's ability to extract features of foreign objects. Experimental results show that compared to the original YOLOv8n model, the proposed model achieves an increase of 8.3% in mean average precision (mAP) at IOU=0.5, reaching 0.957, with a detection speed of 1.5 FPS. This provides a new approach and method for detecting foreign objects in railway catenary systems.

KEYWORDS

Foreign Objects in High-speed Railway Catenary, Deep Learning, YOLOV8n, Characteristic Pyramids, Multi-Scale Attention

CITE THIS PAPER

Jian Sun, Zhenhua Wang, Tianijng Zhang, Siqi Chen, Rui Wang. Research on Foreign Object Detection Algorithm for High-Speed Railway Catenary Based on Improved YOLOv8n. Advances in Computer, Signals and Systems (2026). Vol. 10, No. 2, 1-10. DOI: http://dx.doi.org/10.23977/acss.2026.100201.

REFERENCES

[1] Ren Zhijun, Lin Suzhen, Li Dawei, et al. Mask R-CNN Object Detection Method Based on Improved Feature Pyramid[J].Laser & Optoelectronics Progress, 2019, 56(4):041502.DOI:10.3788/LOP56.041502.
[2] Yin X , Chen L .Image Object Detection Method Based on Improved Faster R-CNN[J].Journal of Circuits, Systems and Computers,2024, 33(07).DOI:10.1142/S0218126624501305.
[3] Sun S , Mo B , Xu J ,et al.Multi-YOLOv8: An infrared moving small object detection model based on YOLOv8 for air vehicle[J].Neurocomputing,2024, 588.DOI:10.1016/j.neucom.2024.127685.
[4] Zhou Jun, Chen Jianyun. Research on Catenary Bird's Nest Recognition and Detection Based on DSSD [J]. Journal of East China Jiaotong University, 2019, 36(6):9.DOI:CNKI:SUN:HDJT.0.2019-06-011.
[5] Wang Xiaohong, Du Yunfei, Liu Chang. Detection of Bird Nests and Foreign Objects in Catenary Systems Based on YOLOv5s [J]. Journal of Yangtze Information and Communication, 2023, 36(6):51-54.
[6] Jiang Xinlan, Jia Wenbo. Machine Vision Detection Methods for Foreign Object Intrusion in High-Speed Railway Catenary Systems [J]. Computer Engineering and Applications, 2019.
[7] Wang Qian. Research on Foreign Object Detection in Railway Catenary Systems Based on Improved YOLOv5 [J]. Academic Research of Xi'an Jiaotong University Institute of Technology, 2023, 8(3):49-53.
[8] Wang Keli, Gao Fulai, Yang Peng, et al. Research on Bird Nest and Foreign Object Recognition in Catenary Systems Based on Deep Learning [J]. Railway Locomotives and Vehicles, 2022(002):042.
[9] He Deqiang, Jiang Zhou, Chen Jiyong, et al. Research on Detection Methods for Bird Nests in Railway Catenary Systems Based on Deep Convolutional Neural Networks [J]. Electric Drive for Locomotives, 2019(4):5.DOI: CNKI:SUN:JCDC.0.2019-04-030.
[10] Wang Jiwu, Luo Haibao, Yu Pengfei, et al. Detection of Bird Nests in Railway Catenary Systems Based on Faster R-CNN [J]. Railway Locomotives and Vehicles, 2020, 40(2):5.DOI:CNKI:SUN:TDJC.0.2020-02-019.
[11] Yang Pei. Research on Dual-Discriminator Generative Adversarial Networks and Their Application in Detecting Bird Nests in Catenary Systems [D]. Southwest Jiaotong University,[2024-07-20].
[12] Wu C , He X , Wu Y .An object detection method for catenary component images based on improved Faster R-CNN[J].IOP Publishing Ltd, 2024.DOI:10.1088/1361-6501/ad4c01.
[13] Kingma D , Ba J .Adam: A Method for Stochastic Optimization[J].Computer Science, 2014.DOI:10.48550/arXiv. 1412.6980.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 46526
Visits: 1058646

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.