Arcing detection method of metro pantograph based on improved yolov4-tiny

AI Shengyong; DING Jianming; ZHANG Qingsong

doi:10.13890/j.issn.1000-128X.2023.04.012

您当前的位置：

首页 >

文章列表页 >

Arcing detection method of metro pantograph based on improved yolov4-tiny

Railway Electrification | 更新时间：2024-08-02

- Arcing detection method of metro pantograph based on improved yolov4-tiny
- Electric Drive for Locomotives Issue 4, Pages: 83-89(2023)
- 作者机构：
  
  西南交通大学轨道交通运载系统全国重点实验室，四川成都 610031
- 作者简介：
- 基金信息：
- DOI：10.13890/j.issn.1000-128X.2023.04.012
  CLC： U231;U225
- Published：10 July 2023，
  
  Received：07 March 2023，
  
  Revised：01 July 2023，
- 稿件说明：
扫描看全文
艾生永, 丁建明, 张青松. 基于改进yolov4-tiny的地铁受电弓燃弧检测方法[J]. 机车电传动, 2023(4): 83-89.

AI Shengyong, DING Jianming, ZHANG Qingsong. Arcing detection method of metro pantograph based on improved yolov4-tiny[J]. Electric drive for locomotives,2023(4): 83-89.
艾生永, 丁建明, 张青松. 基于改进yolov4-tiny的地铁受电弓燃弧检测方法[J]. 机车电传动, 2023(4): 83-89. DOI： 10.13890/j.issn.1000-128X.2023.04.012.

AI Shengyong, DING Jianming, ZHANG Qingsong. Arcing detection method of metro pantograph based on improved yolov4-tiny[J]. Electric drive for locomotives,2023(4): 83-89. DOI： 10.13890/j.issn.1000-128X.2023.04.012.

摘要

地铁受电弓的燃弧会加剧受电弓滑板和接触网磨损，严重危害轨道交通安全。文章针对地铁车辆受电弓燃弧检测存在强光干扰和背景多变的问题，提出了一种基于yolov4-tiny模型改进的燃弧检测方法。为提升小目标检测能力，该方法在yolov4-tiny原有两个尺度的预测分支的基础上，添加第三尺度的预测分支以实现小燃弧在浅层网络中的定位，在主干网络后增加RFB(Receptive Field Block)模块以扩大网络的感受野，增强模型的特征提取能力。结果表明，改进的模型在测试集上的平均精度值（

）比yolov4-tiny提升了7.8个百分点，达到了98.2%，燃弧的定位效果与yolov4相当，但速度得到了极大的提升，单张图片的推理速度仅为6.5 ms，能有效、准确地完成地铁车辆中的受电弓燃弧检测任务。

Abstract

Arcing on metro pantographs contribute to the accelerated wear of pantograph strips and the catenary

imposing a serious safety risk to rail transit. For the problem of strong light interference and variable background in pantograph arcing detection for metro vehicles

an arcing detection method based on an improved yolov4-tiny model is proposed. In order to improve the detection ability of small targets

this method incorporated a third-scale prediction branch to the original two-scale prediction branches of yolov4-tiny

to enable the positioning of small arcs in the shallow network. Moreover

a receptive field block (RFB) module was embedded beneath the backbone network

to expand the network′s receptive field and enhance the model′s feature extraction ability. The test results show the average precision (

) of the improved model on the test set is increased to 98.2%

marking an improvement of 7.8 percentage points over yolov4-tiny. Despite yielding similar positioning effect for arcs

the improved model works at significantly higher speeds than yolov4

enabling inference speed of a single image in only 6.5 ms. The proposed

method is proven effective and accurate in fulfilling the arcing detection task for pantographs of metro vehicles.

关键词

地铁车辆燃弧识别目标检测YOLO深度学习

Keywords

metro vehiclearcing detectiontarget detectionYOLOdeep learning

references

高国强, 郝静, 古圳, 等. 高速铁路中受电弓升弓过程弓网电弧电气特性[J]. 高电压技术, 2016, 42(11): 3569-3575.

GAO Guoqiang, HAO Jing, GU Zhen, et al. Electrical characteristics on pantograph-catenary arc during pantograph rising in high speed railway[J]. High voltage engineering, 2016, 42(11): 3569-3575.

KARAKOSE E, GENCOGLU M T, KARAKOSE M, et al. A new arc detection method based on fuzzy logic using S-transform for pantograph-catenary systems[J]. Journal of intelligent manufacturing, 2018, 29(4): 839-856.

LI Bin, LUO Chenyu, WANG Zhiyong. Application of GWO-SVM algorithm in arc detection of pantograph[J]. IEEE access, 2020, 8: 173865-173873.

王智勇, 郭凤仪, 冯晓丽, 等. 基于电流信号特征的弓网电弧识别方法[J]. 电工技术学报, 2018, 33(1): 82-91.

WANG Zhiyong, GUO Fengyi, FENG Xiaoli, et al. Recognition method of pantograph arc based on current signal characteristics[J]. Transactions of China electrotechnical society, 2018, 33(1): 82-91.

刘宝轩, 陈唐龙, 于龙, 等. 地铁弓网燃弧能量检测与牵引电流扰动分析[J]. 铁道学报, 2015, 37(3): 8-13.

LIU Baoxuan, CHEN Tanglong, YU Long, et al. Pantograph-catenary arc energy detection and analysis on current disturbance in metro[J]. Journal of the China railway society, 2015, 37(3): 8-13.

于晓英, 苏宏升. 基于PMT电压一次积分值的城轨弓网电弧检测系统[J]. 铁道学报, 2019, 41(9): 51-58.

YU Xiaoying, SU Hongsheng. Arcing detection system for pantograph-catenary system of urban rail transit based on voltage integral value of PMT[J]. Journal of the China railway society, 2019, 41(9): 51-58.

AYDIN I, KARAKOSE M, AKIN E. Anomaly detection using a modified kernel-based tracking in the pantograph-catenary system[J]. Expert systems with applications, 2015, 42(2): 938-948.

AYDIN I. A new approach based on firefly algorithm for vision-based railway overhead inspection system[J]. Measurement, 2015, 74: 43-55.

杨恒, 伍川辉, 吴琛. 基于图像处理弓网燃弧检测研究[J]. 铁道科学与工程学报, 2018, 15(4): 1030-1035.

YANG Heng, WU Chuanhui, WU Chen. The pantograph-catenary arc test research based on image processing technology[J]. Journal of railway science and engineering, 2018, 15(4): 1030-1035.

LU Poyan, HUO Changfan, DUAN Wangwang, et al. Information fusion and image processing based arc detection and localization in pantograph-catenary systems[C]//IEEE. 2019 22th International Conference on Information Fusion (FUSION). Ottawa: IEEE, 2019: 1-8.

XING Zongyi, ZHANG Zhenyu, YAO Xiaowen, et al. Rail wheel tread defect detection using improved YOLOv3[J]. Measurement, 2022, 203: 111959.

WU Yunpeng, QIN Yong, QIAN Yu, et al. Automatic detection of arbitrarily oriented fastener defect in high-speed railway[J]. Automation in construction, 2021, 131: 103913.

YAO Zikai, HE Deqiang, CHEN Yanjun, et al. Inspection of exterior substance on high-speed train bottom based on improved deep learning method[J]. Measurement, 2020, 163: 108013.

HE Deqiang, ZOU Zhiheng, CHEN Yanjun, et al. Obstacle detection of rail transit based on deep learning[J]. Measurement, 2021, 176: 109241.

GUO Feng, QIAN Yu, SHI Yuefeng. Real-time railroad track components inspection based on the improved YOLOv4 framework[J]. Automation in construction, 2021, 125: 103596.

LIU Songtao, HUANG Di, WANG Yunhong. Receptive field block net for accurate and fast object detection[C]//Springer. Computer Vision-ECCV 2018. Switzerland: Springer, 2018: 404-419.

WANG C Y, BOCHKOVSKIY A, LIAO H Y M. Scaled-YOLOv4: scaling cross stage partial network[C]//IEEE. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE, 2021: 13024-13033.

BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[DB/OL]. (2020-04-23) [2023-01-16]. https://arxiv.org/abs/2004.10934https://arxiv.org/abs/2004.10934.

SZEGEDY C, LIU Wei, JIA Yangqing, et al. Going deeper with convolutions[C]//IEEE. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE, 2015: 1-9.

HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//IEEE. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 770-778.

Views

下载量

CSCD

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

Pedestrian detection method in rail transit scenes based on fusion of 3D point clouds and images

A virtual rail train path recognition method based on binocular stereo vision

Lightweight YOLO models for object detection based on low-rank decomposition

A study on life assessment of metro cowcatcher based on multi-axis vibration environment reconstruction

An end-to-end text-to-speech system for vehicle-mounted devices

Related Author

HE Jia

ZENG Junwei

GENG Li

LU Chenyang

REN Lihui

LIN Delü

LIU Chang

CHEN Qi

Related Institution

CHN ENERGY Baoshen Railway Co., Ltd.

Institute of Rail Transit, Tongji University

School of Computer Science and Technology, Huazhong University of Science and Technology

CRRC Zhuzhou Institute Co., Ltd.

School of Mechanical Engineering, Southwest Jiaotong University

⁰