多传感器信息融合的铁路扣件缺陷检测方法

doi:10.3901/JME.2021.20.038

机械工程学报 ›› 2021, Vol. 57 ›› Issue (20): 38-46.doi: 10.3901/JME.2021.20.038

扫码分享

多传感器信息融合的铁路扣件缺陷检测方法

金鹏¹, 黄浩¹, 刘检华¹, 刘少丽¹, 方玥², 何森¹, 戚慧志¹, 刘威¹

1. 北京理工大学机械与车辆学院北京 100081;
2. 中国铁道科学研究院集团有限公司基础设施检测研究所北京 100081

收稿日期:2020-12-15 修回日期:2021-05-16 出版日期:2021-10-20 发布日期:2021-12-15
通讯作者: 刘少丽(通信作者),女,1984年出生,博士,副教授,博士研究生导师。主要研究方向为数字化检测与机器视觉。E-mail:liushaoli@bit.edu.cn
作者简介:金鹏,男,1987年出生,博士后。主要研究方向为物体三维重建技术。E-mail:kingpeng604@163.com;方玥,女,1984年出生,博士,副研究员。主要研究方向为轨道检测与机器视觉。E-mail:fangyue@rails.cn
基金资助:
国家自然科学基金（51875044）和国家基础科研（JCKY2017204B502）资助项目。

Fault Detection Method of Railway Fastener Combined with Multi-sensor Information

JIN Peng¹, HUANG Hao¹, LIU Jianhua¹, LIU Shaoli¹, FANG Yue², HE Sen¹, QI Huizhi¹, LIU Wei¹

1. School of Mechanical Engineering, Beijing Institute of Technology, Beijing 100081;
2. Infrastructure Inspection Research Institute, China Academy of Railway Sciences Corporation Limited, Beijing 100081

Received:2020-12-15 Revised:2021-05-16 Online:2021-10-20 Published:2021-12-15

摘要/Abstract

摘要： 基于机器视觉的二维图像铁路扣件缺陷检测已经取代人工检测，提高了检测效率。但是，铁路扣件缺陷样本数量少且标注困难，以及检测结果受光照条件影响大等问题仍然是当前所面对的主要挑战。因此，提出一种多传感器信息融合的铁路扣件缺陷检测方法，采用结构光设备快速高效地采集铁路扣件的深度和强度信息，设计一种针对深度图的自适应定位分割方法，自动准确定位和分割深度图中的铁路扣件，并将定位分割参数与强度图融合，实现铁路扣件的定位分割；针对样本不均衡和样本标注效率低的问题，通过DCGAN扩充缺陷扣件样本的数量；设计了具有8个残差块共18层的ResNet神经网络，完成铁路扣件螺母缺失、弹条断裂和弹条缺失三种缺陷类别检测，并采集WJ-7类型铁路扣件样本设计了相关试验，验证结果表明，缺陷扣件检测平均准确率达到97.6%，满足工程应用的需求。

关键词: 铁路扣件, 缺陷检测, 样本不平衡, 深度学习

Abstract: The defect detection of railway fastener based on machine vision has replaced manual detection and improved the detection efficiency. However, the problems including few samples of defective railway fastener, the difficulty of sample labeling and the great influence of light conditions on 2D image detection are still the main obstacles in defect detection of railway fastener. Therefore, a defective railway fastener detection method combining multi-sensor information is proposed. The sample of railway fastener is quickly and efficiently collected by structural light equipment. An adaptive positioning and segmentation method based on depth map was designed to automatically and accurately locate and segment railway fastener in depth map, and the positioning and segmentation parameters were integrated to realize the positioning and segmentation of railway fastener on the intensive map. To solve the problem of unbalanced samples and low efficiency of sample labeling, DCGAN is used to expand the number of samples of defective railway fastener. In addition, a depth neural network with a total of 18 layers of 8 residual blocks was designed based on ResNet network to complete the detection of three types of defects:missing nut, broken spring strip and missing spring strip of railway fastener, and samples of WJ-7 type railway fastener were collected for experiment to verify the proposed method. The results showed that the average accuracy of defect detection of fastener was 97.6%, meeting the needs of engineering applications.

Key words: railway fasten, fault detection, imbalance sample, deep learning

中图分类号:

U216

金鹏, 黄浩, 刘检华, 刘少丽, 方玥, 何森, 戚慧志, 刘威. 多传感器信息融合的铁路扣件缺陷检测方法[J]. 机械工程学报, 2021, 57(20): 38-46.

JIN Peng, HUANG Hao, LIU Jianhua, LIU Shaoli, FANG Yue, HE Sen, QI Huizhi, LIU Wei. Fault Detection Method of Railway Fastener Combined with Multi-sensor Information[J]. Journal of Mechanical Engineering, 2021, 57(20): 38-46.

参考文献

[1] 肖新标,金学松,温泽峰. 钢轨扣件失效对列车动态脱轨的影响[J]. 交通运输工程学报,2006,6(1):10-15. XIAO Xinbiao,JIN Xuesong,WEN Zefeng. Influence of rail fastener failure on vehicle dynamic derailment[J]. Journal of the Traffic and Transportation Engineering, 2006,6(1):10-15.
[2] ZHOU L,SHEN Z Y. Dynamic analysis of a high-speed train operating on a curved track with failed fasteners[J]. Journal of Zhejiang University:Science A,2013,14(6):447-458.
[3] XIAO X,JIN X,WEN Z. Effect of disabled fastening systems and ballast on vehicle derailment[J]. Journal of Vibration and Acoustics,2006,129(2):217-229.
[4] MORALES-IVORRA S,REAL J I,HERNÁNDEZ C,et al. Derailment risk and dynamics of railway vehicles in curved tracks:Analysis of the effect of failed fasteners[J]. Journal of Modern Transportation,2016,24(1):38-47.
[5] ESVELD C. Modern railway track[M]. Zaltbommel:MRT-productions,2001.
[6] MAO Q,CUI H,HU Q,et al. A rigorous fastener inspection approach for high-speed railway from structured light sensors[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2018,143:249-267.
[7] 代先星,阳恩慧,WANG K C,等. 铁路扣件缺陷自动检测研究进展[J]. 铁道科学与工程学报,2017,14(11):2452-2459. DAI Xianxing,YANG Enhui,WANG K C,et al. Technology progress of automatic railway fastener inspection[J]. Journal of Railway Science and Engineering,2017,14(11):2452-2459.
[8] SHOLL H,AMMAR R,GREENSHIELDS I,et al. Application of computing analysis to real-time railroad track inspection[C/CD]//2006 World Automation Congress,24-26 Jul. 2006,Budapest,Hungary,2006.
[9] SINGH M,SINGH S,JAISWAL J,et al. Autonomous rail track inspection using vision based system[C/CD]//2006 IEEE International Conference on Computational Intelligence for Homeland Security and Personal Safety,16-17 Oct. 2006,Alexandria,VA,USA,2006.
[10] XIE Y,TALBOT M,ZHAO K,et al. Automatic track inspection using 3D laser profilers to improve rail transit asset condition assessment and state of good repair:A preliminary study[C/CD]//TRB 2014 Annual Meeting,2014.
[11] 许贵阳,史天运,任盛伟,等. 基于计算机视觉的车载轨道巡检系统研制[J]. 中国铁道科学,2013(1):139-144. XU Guiyang,SHI Tianyun,REN Shengwei,et al. Development of the on-board track inspection system based on computer vision[J]. China Railway Science, 2013(1):139-144.
[12] MARINO F,DISTANTE A,MAZZEO P L,et al. A real-time visual inspection system for railway maintenance:Automatic hexagonal-headed bolts detection[J]. IEEE Transactions on Systems Man & Cybernetics Part C,2007,37(3):418-428.
[13] RESENDIZ E,HART J M,AHUJA N. Automated visual inspection of railroad tracks[J]. IEEE Transactions on Intelligent Transportation Systems,2013,14(2):751-760.
[14] XIA Y,XIE F,JIANG Z. Broken railway fastener detection based on adaboost algorithm[C]//International Conference on Optoelectronics and Image Processing,11-12 Nov. 2010,Haikou,China,2010.
[15] FENG H,JIANG Z,XIE F,et al. Automatic fastener classification and defect detection in vision-based railway inspection systems[J]. IEEE Transactions on Instrumentation & Measurement,2014,63(4):877-888.
[16] GIBERT X,PATEL V M,CHELLAPPA R. Robust fastener detection for autonomous visual railway track inspection[C/CD]//IEEE Winter Conference on Applications of Computer Vision,5-9 Jan,2015,Waikoloa,HI,USA,2015.
[17] FAN H,PAMELA C C,YUN H,et al. High speed railway fastener detection based on line local binary pattern[J]. IEEE Signal Processing Letters,2018,25(6):788-792.
[18] 代先星,阳恩慧,丁世海,等. 基于三维图像的铁路扣件缺陷自动识别算法[J]. 铁道学报,2017,39(10):89-96. DAI Xianxing,YANG Enhui,DING Shihai,et al. Automatic defect inspection algorithm of railway fasteners based on 3D images[J]. Journal of the China Railway Society,2017,39(10):89-96.
[19] AYTEKIN A,REZAEITABAR Y,DOGRU S,et al. Railway fastener inspection by real-time machine vision[J]. IEEE Transactions on Systems,Man,and Cybernetics:Systems, 2017,45(7):1101-1107.
[20] LORENTE A G,LLORCA D F,VELASCO M G,et al. Range-based rail gauge and rail fasteners detection using high-resolution 2D/3D images[C]//TRB 2014 Annual Meeting. TRB2014,January 12-16,Washington,D.C.,United States,2014:125-132.
[21] CUI H,HU Q,MAO Q. Real-time geometric parameter measurement of high-speed railway fastener based on point cloud from structured light sensors[J]. Sensors,2018,18(11):3675.
[22] HAN Q,WANG S,FANG Y,et al. A rail fastener tightness detection approach using multi-source visual sensor[J]. Sensors,2020,20(5):1367.
[23] SHORTEN C,KHOSHGOFTAAR T M. A survey on image data augmentation for deep learning[J]. Journal of Big Data,2019,6:60.
[24] GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al. Generative adversarial nets[C]//Advances in neural information processing systems,NIPS2014,December 8-13,Montreal,Quebec,Canada,2014:2672-2680.
[25] CRESWELL A,WHITE T,DUMOULIN V,et al. Generative adversarial networks:An overview[J]. IEEE Signal Processing Magazine,2018,35(1):53-65.
[26] LEE S,PARK W,JUNG S,et al. Fault detection of aircraft system with random forest algorithm and similarity measure[J]. Scientific World Journal,2014,6:727359.
[27] LIU H,ZHOU J,XU Y,et al. Unsupervised fault diagnosis of rolling bearings using a deep neural network based on generative adversarial networks[J]. Neurocomputing,2018,315(13):412-424.
[28] ZAREAPOOR M,SHAMSOLMOALI P,YANG J. Oversampling adversarial network for class-imbalanced fault diagnosis[J]. Mechanical Systems and Signal Processing,2021,149:107175.

多传感器信息融合的铁路扣件缺陷检测方法

Fault Detection Method of Railway Fastener Combined with Multi-sensor Information

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	李春凯, 王嘉昕, 石玗, 代悦. GTAW熔池激光条纹动态行为与熔透预测方法研究[J]. 机械工程学报, 2024, 60(6): 236-244.
[2]	傅杨, 张跃, 毛颖, 唐小华, 陈祖高, 徐和武, 杨雨沛, 高斌, 田贵云. 基于Feature Boosting的管道电磁内检测多传感信号缺陷解析算法[J]. 机械工程学报, 2024, 60(20): 51-67.
[3]	赵德望, 姜超, 赵延广, 杨文平, 樊俊铃. 铝锂合金疲劳性能及预测研究-基于试验和“浅层”+“深度”混合神经网络方法[J]. 机械工程学报, 2024, 60(16): 190-199.
[4]	付玲, 佘玲娟, 颜镀镭, 张鹏, 龙湘云. 基于内嵌物理信息与注意力机制BiLSTM神经网络的臂架系统疲劳损伤预测模型[J]. 机械工程学报, 2024, 60(13): 205-215.
[5]	陈钱, 陈康康, 董兴建, 皇甫一樊, 彭志科, 孟光. 一种面向机械设备故障诊断的可解释卷积神经网络[J]. 机械工程学报, 2024, 60(12): 65-76.
[6]	刘岳开, 王天杨, 褚福磊. 基于低延迟可解释性深度学习的复杂旋转机械关键部件知识嵌入与诊断方法研究[J]. 机械工程学报, 2024, 60(12): 107-115.
[7]	赖旭伟, 丁昆, 张楷, 黄锋飞, 郑庆, 李致萱, 丁国富. 基于可解释物理引导空间注意力改进的跨工艺参数立铣刀磨损辨识[J]. 机械工程学报, 2024, 60(12): 147-157.
[8]	林昕, 毛杨坤, 傅盈西, 朱锟鹏. 基于图像语义分割与图卷积的选区激光熔融成形过程羽流运动特征分析[J]. 机械工程学报, 2024, 60(12): 168-182.
[9]	邵海东, 肖一鸣, 邓乾旺, 任颖莹, 韩特. 基于不确定性感知网络的可信机械故障诊断[J]. 机械工程学报, 2024, 60(12): 194-206.
[10]	李玖法, 邹博文, 任玥. 考虑车-路交互作用的车辆轨迹预测算法研究[J]. 机械工程学报, 2024, 60(10): 76-85.
[11]	许思远, 张卫东, 毛玉明, 牛斌. 数据驱动的指定变形非均质多胞结构优化设计[J]. 机械工程学报, 2024, 60(1): 85-95.
[12]	郎宁, 王德成, 程鹏. 基于集成自适应欠采样的铝管表面缺陷检测方法研究[J]. 机械工程学报, 2023, 59(6): 18-31.
[13]	何赟泽, 李响, 王洪金, 侯岳骏, 张帆, 牟欣颖, 刘浩, 程豪, 李时华, 李杰. 基于可见光和热成像的风机叶片全周期无损检测综述[J]. 机械工程学报, 2023, 59(6): 32-45.
[14]	陆向宁, 刘凡, 何贞志, 廖广兰, 史铁林. 稀疏重构SAM芯片焊点检测方法研究[J]. 机械工程学报, 2023, 59(6): 95-102.
[15]	宿磊, 王立建, 祁阳, 张思雨, 顾杰斐, 李可. 基于IADSA深度迁移网络的金属表面缺陷检测[J]. 机械工程学报, 2023, 59(24): 46-55.