复杂天气下铁轨异物入侵目标检测研究

魏雨桐; 侯涛; 牛宏侠

doi:10.7652/xjtuxb202610010

您当前的位置：

首页 >

文章列表页 >

复杂天气下铁轨异物入侵目标检测研究

专题智能感知与自主决策 | 更新时间：2026-09-29

- 复杂天气下铁轨异物入侵目标检测研究
- Research on Object Detection for Foreign Object Intrusion on Railway Tracks under Complex Weather Conditions
- 西安交通大学学报 2026年60卷第10期页码：113-124
- 作者机构：
  
  兰州交通大学自动化与电气工程学院，730070，兰州
- 作者简介：
  
  魏雨桐（2003-），女，硕士生；
  侯涛（通信作者），男，教授，硕士生导师。
- 基金信息：
  
  甘肃省重点研发计划资助项目。(23YFGA0049)
- DOI：10.7652/xjtuxb202610010
  中图分类号： TP391.41
- 收稿：2026-03-10，
  
  网络首发：2026-05-07，
  
  纸质出版：2026-10-10
- 稿件说明：
移动端阅览
魏雨桐, 侯涛, 牛宏侠. 复杂天气下铁轨异物入侵目标检测研究[J/OL]. 西安交通大学学报,2026,60 (10):113-124. https://doi.org/10.7652/xjtuxb202610010.

WEI Yutong, HOU Tao, NIU Hongxia. Research on Object Detection for Foreign Object Intrusion on Railway Tracks under Complex Weather Conditions[J/OL]. Journal of Xi'an Jiaotong University,2026,60 (10):113-124. https://doi.org/10.7652/xjtuxb202610010.
魏雨桐, 侯涛, 牛宏侠. 复杂天气下铁轨异物入侵目标检测研究[J/OL]. 西安交通大学学报,2026,60 (10):113-124. https://doi.org/10.7652/xjtuxb202610010. DOI：

WEI Yutong, HOU Tao, NIU Hongxia. Research on Object Detection for Foreign Object Intrusion on Railway Tracks under Complex Weather Conditions[J/OL]. Journal of Xi'an Jiaotong University,2026,60 (10):113-124. https://doi.org/10.7652/xjtuxb202610010. DOI：

摘要

为保障复杂天气下列车运行安全，针对复杂天气下，铁轨异物入侵目标检测算法参数量大、清晰度低、易出现误检漏检及推理速度慢等问题，提出基于改进YOLOv11n算法的复杂天气铁路轨道异物入侵检测算法（RCW-YOLO）。首先，在YOLOv11n算法基础上新增雨雾感知动态融合门控，并设计了雨雾浓度估计器模块；然后，在目标检测头框架DynamicHead中新增雨雾注意力分支，并在卷积偏移量中融入雨雾浓度系数，减少误检率和漏检率；其次，在骨干和颈部网络中使用注意力增强多子带融合小波池化模块替代YOLOv11n现有卷积模块，在保持检测效果的同时减少参数量，提高推理速度；最后，引入新设计的损失函数EW-MPDIoU替代原损失函数CIOU，提高检测精度。实验结果表明：采用所提RCW-YOLO算法的平均检测精度为84.6%，较YOLOv11n算法提升了9.2%，参数量下降了18.15%；与主流目标检测算法YOLOv9t和YOLOv10n相比，采用RCW-YOLO算法的平均检测精度分别提升了11.3%和11.3%，参数量分别下降了21.63%、25.34%；相比最新的YOLOv12和YOLOv13算法，平均检测精度分别提高了7.8%、7.0%，参数量分别下降了22.18%和19.93%。

Abstract

To ensure the safety of train operations under complex weather conditions

and to address the issues of existing object detection algorithms for foreign object intrusion on railway tracks

such as a large number of parameters

poor image clarity

high rates of false and missed detections

and slow inference speed

a foreign object intrusion detection algorithm for railway tracks under complex weather conditions (RCW-YOLO) based on the improved YOLOv11n is proposed. First

based on the YOLOv11n algorithm

a rain-fog perception dynamic fusion gating mechanism is introduced

and a rain-fog density estimator module is designed. Second

a rain-fog attention branch is added to the object detection head framework DynamicHead

and a rain-fog density coefficient is integrated into the convolutional offset to reduce the false and missed detection rates. Third

in the backbone and neck networks

an attention-enhanced multi-subband fusion wavelet pooling module is utilized to replace the existing convolutional modules of YOLOv11n

which reduces the number of parameters and improves inference speed while maintaining detection performance. Finally

a newly designed loss function

EW-MPDIoU

is introduced to replace the original CIoU loss function

thereby improving detection accuracy. Experimental results show that the proposed RCW-YOLO algorithm achieves an average detection accuracy of 84.6%

which is 9.2% higher than that of the YOLOv11n algorithm

while the number of parameters is reduced by 18.15%. Compared with the mainstream object detection algorithms YOLOv9t and YOLOv10n

the average detection accuracy of the RCW-YOLO algorithm is improved by 11.3% and 11.3%

respectively

and the number of parameters is reduced by 21.63% and 25.34%

respectively. Furthermore

compared with the latest YOLOv12and YOLOv13algorithms

the average detection accuracy is improved by 7.8% and 7.0%

respectively

and the number of parameters is reduced by 22.18% and 19.93%

respectively.

关键词

Keywords

references

徐展.基于轻量化目标检测的铁路周界异物入侵检测方法研究[D].沈阳: 沈阳工业大学, 2025.

赵永超, 丁东, 王汉策, 等.基于无人值守视觉传感系统的铁路边坡侵限落石智能监测方法[J].科技与创新, 2025, (20): 77-79.

Zhao Yongchao, Ding Dong, Wang Hance, et al. Intelligent monitoring method for railway slope encroaching rockfall based on unattended visual sensing system[J].Science and Innovation, 2025, (20): 77-79.

Hou Pengfei, Chen Li.Rain-fog detection via spatial adaptive deformable network with multi-scale feature preservation and task-aware dynamic calibration [J]. Engineering Research Express, 2025, 7(3): 035248.

朱思远.基于自适应变化检测和自监督机制的异物侵限检测方法研究[D].西安: 西安交通大学, 2022.

杨伟, 王帅, 吴佳奇, 等.从局部到全局的零参考低照度图像增强方法[J].西安交通大学学报, 2024, 58(4): 158-169.

Yang Wei, Wang Shuai, Wu Jiaqi, et al. A zero-reference low-light image enhancement method from local to global[J].Journal of Xi'an Jiaotong University, 2024, 58(4): 158-169.

高修强, 余星阳, 刘伯鹍, 等.基于改进Mask R-CNN的轨道交通异物入侵模型研究[J].自动化与仪表, 2025, 40(6): 111-115.

Gao Xiuqiang, Yu Xingyang, Liu Bokun, et al. Research on foreign object intrusion model for rail transit based on im-proved maskR-CNN[J].Automation &Instrumentation, 2025, 40(6): 111-115.

陈俊明.面向铁路异物入侵检测的图像数据扩增方法研究[D].北京: 北京交通大学, 2023.

柳长源, 夏苏辉, 兰朝凤.恶劣天气下的航拍车辆目标检测算法[J/OL].北京航空航天大学学报. (2025-12-25)[2026-01-14]. https://doi.org/10.13700/j. bh.1001-5965.2025.0471.

Liu Changyuan, Xia Suhui, Lan Chaofeng.Algorithm for aerial vehicle object detection under adverse weather[J/OL].Journal of Beijing University of Aeronautics and Astronautics. (2025-12-25)[2026-01-14]. https://doi.org/10.13700/j.bh.1001-5965.2025.0471

王诚, 谢立云, 李坤.基于改进YOLOv11的无人机雾天车辆目标检测算法研究[J/OL].南京邮电大学学报(自然科学版). (2026-03-09)[2026-03-11]. https://link.cnki.net/urlid/32.1772.TN.20260307.1941.008.

Wang Cheng, Xie Liyun, Li Kun.Research on UAV vehicle target detection algorithm in foggy weather based on improved YOLOv11[J/OL]. Journal of Nanjing University of Posts and Telecommunications (Natural Science Edition). (2026-03-09)[2026-03-11]. https: //link.cnki.net/urlid/32.1772.TN.20260307.1941.008.

王井阳, 徐勇超, 张波, 等.改进YOLOv10n的轻量化道路裂缝检测模型[J/OL].郑州大学学报(工学版). (2026-02-28)[2026-03-11]. https://doi.org/10.13705/j.issn.1671-6833.2026.04.007.

Wang Jingyang, Xu Yongchao, Zhang Bo, et al. Lightweight road crack detection model based on improved YOLOv10n[J/OL].Journal of Zhengzhou University (Engineering Science). (2026-02-28)[2026-03-11]. https://doi.org/10.13705/j.issn.1671-6833.2026.04.007.

郭家虎, 何磊.DEL-YOLO: 低照度轻量级煤矿输送带异物检测[J].电子测量与仪器学报, 2025, 39(12): 289-299.

Guo Jiahu, He Lei.DEL-YOLO: low-illumination lightweight object detection for conveyor belts in coalmines[J].Journal of Electronic Measurement and Instrumentation, 2025, 39(12): 289-299.

孙修宇, 毕楠.基于改进YOLO11的低光照环境目标检测算法[J].激光与光电子学进展, 2025, 62(22): 417-427.

Sun Xiuyu, Bi Nan.Low-light environment object detection algorithm based on improvedYOLO11[J].Laser & Optoelectronics Progress, 2025, 62(22): 417-427.

Ruan Chunliang, Hong Liang, Gao Junjie.Marine-YOLO: a high-precision object detection algorithm for complex maritime environments [J].Applied Ocean Research, 2026, 169: 104994.

张瑞芳, 段世纪, 刘占占, 等.MSDE-YOLO: 低照度图像多尺度特征聚合目标检测算法[J/OL].电光与控制. (2026-01-09)[2026-01-14]. https://link.cnki. net/urlid/41.1227.tn.20260108.1603.004.

Zhang Ruifang, Duan Shiji, Liu Zhanzhan, et al. MSDE-YOLO: multi-scale feature aggregation for target detection in Low-Illumination images[J/OL].Electronics Optics & Control. (2026-01-09)[2026-01-14]. https://link.cnki.net/urlid/41.1227.tn.20260108.1603.004.

周强.低光图像目标检测的优化研究[J/OL].智能计算机与应用. (2025-12-31)[2026-01-14]. https://doi.org/10.20169/j.issn.2095-2163.25120401.

Zhou Qiang.Research on optimization of object detection in low-light images[J/OL].Intelligent Computer and Applications. (2025-12-31)[2026-01-14]. https://doi.org/10.20169/j.issn.2095-2163.25120401.

王轩楷, 潘广贞, 陈新.改进YOLO11n的水下目标检测[J].计算机系统应用, 2025, 34(10): 76-85.

Wang Xuankai, Pan Guangzhen, Chen Xin.Improved underwater target detection with YOLO11n[J].Computer Systems & Applications, 2025, 34(10): 76-85.

牛为华, 郭迅.融合分块注意力与小波特征聚合的遥感图像目标检测算法[J/OL].智能系统学报. (2026-03-09)[2026-05-06]. https://link.cnki.net/urlid/23.1538.tp.20260306.1518.004.

Niu Weihua, Guo Xun.Remote sensing object detection algorithm integrating block attention and wavelet feature aggregation[J/OL].CAAI Transactions on Intelligent Systems. (2026-03-09)[2026-05-06]. https://link.cnki.net/urlid/23.1538.tp.20260306.1518.004.

王楠, 侯涛, 牛宏侠.多尺度特征融合的铁轨异物入侵检测研究[J].西安交通大学学报, 2024, 58(9): 139-153.

Wang Nan, Hou Tao, Niu Hongxia.Research on railway track foreign object intrusion detection based on multi-scale feature fusion[J].Journal of Xi'an Jiaotong University, 2024, 58(9): 139-153.

张凡, 张鹏超, 王磊, 等.基于YOLOv5s的轻量化朱鹮检测算法研究[J].西安交通大学学报, 2023, 57(1): 110-121.

Zhang Fan, Zhang Pengchao, Wang Lei, et al. Research on lightweight crested ibis detection algorithm based on YOLOv5s[J].Journal of Xi'an Jiaotong University, 2023, 57(1): 110-121.

侯涛, 席超, 牛宏侠.复杂环境下铁路轨道异物入侵轻量化检测研究[J].北京交通大学学报, 2026, 50(3): 164-174.

Hou Tao, Xi Chao, Niu Hongxia.Research on lightweight detection of foreign objects intrusion on railway tracks with complex environments[J].Journal of Beijing Jiaotong University, 2026, 50(3): 164-174.

郝芝建, 牟舵, 王留毅, 等.基于改进YOLOv10的隧道渗漏水智能检测方法[J/OL].华南理工大学学报(自然科学版). (2026-04-13)[2026-04-15]. https://link. cnki.net/urlid/44.1251.T.20260411.1642.002.

Hao Zhijian, Mu Duo, Wang Liuyi, et al. Intelligent detection method for tunnel leakage water based on improved YOLOv10[J/OL].Journal of South China University of Technology (Natural Science Edition). (2026-04-13)[2026-04-15]. https://link.cnki.net/urlid/44.1251.T. 20260411.1642.002.

刘晓雅, 许贵阳, 白堂博, 等.基于YOLO11-DFI的铁路沿线隐患物智能检测算法[J/OL].铁道建筑. (2026-01-05)[2026-01-14]. https://link.cnki.net/urlid/11.2027.U.20260401.0853.002.

Liu Xiaoya, Xu Guiyang, Bai Tangbo, et al. Intelligent detection algorithm for hidden hazards along railway lines based onYOLO11-DFI[J/OL].Railway Engineering. (2026-01-05)[2026-01-14]. https://link.cnki.net/urlid/11.2027.U.20260401.0853.002.

张红瑞, 冯威铭, 杨潞霞, 等.基于YOLO11改进的水下小目标检测算法CSAF-YOLO [J/OL].计算机应用. (2026-01-08)[2026-01-14]. https://link.cnki.net/urlid/51.1307.TP.20260108.1256.004.

Zhang Hongrui, Feng Weiming, Yang Luxia, et al. CSAF-YOLO: improved YOLO11algorithm for underwater small object detection [J/OL].Journal of Computer Applications. (2026-01-08) [2026-01-14]. https://link.cnki.net/urlid/51.1307.TP.20260108.1256.004.

张涛, 田秀华, 沈健, 等.基于Rail-YOLO的铁路限界入侵人员检测算法研究[J].铁道通信信号, 2026, 62(4): 34-42.

Zhang Tao, Tian Xiuhua, Shen Jian, et al. Study on railway boundary intrusion personnel detection algorithm based on Rail-YOLO[J].Railway Signalling &Communication, 2026, 62(4): 34-42.

Lin T Y, Maire M, Belongie S, et al. Microsoft CO-CO: common objects in context [C]//Computer Vision-ECCV 2014.Cham, Germany: Springer International Publishing, 2014: 740-755.

Everingham M, Eslami S M A, Van Gool L, et al. The Pascal visual object classes challenge: a retrospective [J].International Journal of Computer Vision, 2015, 111(1): 98-136.

浏览量

106

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用置信度分层协同决策的目标检测方法

利用MPEG-4视频流的运动目标检测

一种高分辨率遥感图像视感知目标检测算法

一种改进的单步多框目标检测算法

厌氧消化1号模型在光发酵制氢过程的应用与优化