基于深度学习算法的铁路列车运行安全检测*

doi:10.16265/j.cnki.issn1003-3033.2018.S2.008

中国安全科学学报 ›› 2018, Vol. 28 ›› Issue (S2): 41-45.doi: 10.16265/j.cnki.issn1003-3033.2018.S2.008

基于深度学习算法的铁路列车运行安全检测^*

汪洋¹ 高级工程师, 王俊刚² 提高待遇高工

1 国家铁路局规划与标准研究院,北京 100055;
2 铁路总公司中国铁路北京局集团有限公司,北京 100860

收稿日期:2018-09-19 修回日期:2018-11-13 出版日期:2018-12-30 发布日期:2020-11-11
作者简介:汪洋 (1975—),男,浙江金华人,硕士,高级工程师,主要从事交通运输方面的工作。E-mail:13661380301@139.com。
基金资助:
中国铁路总公司科技研究开发计划项目(2017X014-A)。

Study on safety inspection of railway train operation based on deep learning algorithm

WANG Yang¹, WANG Jungang²

1 Planning and Standard Research Institute, National Railway Administration, Beijing 100055, China;
2 China Railway Beijing Group Co., Ltd., China Railway Corporation, Beijing 100860, China

Received:2018-09-19 Revised:2018-11-13 Online:2018-12-30 Published:2020-11-11

摘要/Abstract

摘要： 随着铁路列车运行安全要求的不断提高,迫切需要及时检测并处理列车运行中的安全隐患。传统的铁路列车运行自动化检测手段主要基于机器视觉算法,存在对列车运行的复杂自然场景适应能力差、误报率高等问题。为解决传统方法检测精度较低等问题,模拟多层神经网络传递结构,输入和识别隐患扫描素材,提出基于深度学习算法的列车运行安全检测方法;然后以铁路货车异物检测为例进行仿真试验,验证相关算法。结果表明:新检测方法在保证检出率的前提下降低了误报率。

关键词: 列车, 安全检测, 深度学习, 混淆矩阵, 迭代

Abstract: With the continuous improvement of railway train operation safety requirements, it is urgent to detect and deal with hidden dangers in train operation in time. The traditional automatic detection method of railway train operation was mainly based on machine vision algorithm, which had the problems of poor adaptability to complex natural scenes and high false alarm rate. In order to solve the problem of low detection accuracy of traditional methods, multi-layer neural network transmission structure was simulated, where hidden trouble scanning materials were input and identified, and a train operation safety detection method based on deep learning algorithm was proposed. Then, a simulation experiment is carried out to verify the relevant algorithm by taking the detection of foreign matter as an example. The results show that the new detection method can reduce the false alarm rate under the premise of ensuring the detection rate.

Key words: train, safety inspection, deep learning, confusion matrix, iteration

中图分类号:

X924.2

汪洋, 王俊刚. 基于深度学习算法的铁路列车运行安全检测^*[J]. 中国安全科学学报, 2018, 28(S2): 41-45.

WANG Yang, WANG Jungang. Study on safety inspection of railway train operation based on deep learning algorithm[J]. China Safety Science Journal, 2018, 28(S2): 41-45.

参考文献

[1] MACQUEEN J. Some methods for classification and analysis of multivariate observations[C]. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics & Probability, 1965:281-297.
[2] WELLS K, BRADLEY D A. A review of X-ray explosives detection techniques for checked baggage[J]. Applied Radiation & Isotopes, 2012, 70(8): 1 729-1 746.
[3] GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning[M].Cambridge: The MIT Press, 2016:68-95.
[4] 朱海港.基于深度学习的高清航拍图像目标检测[D].北京:中国科学院大学,2015.
ZHU Haigang. Research on high resolution aerial objection based on deep learning[D]. Beijing: University of Chinese Academy of Sciences, 2015.
[5] 黄峰,徐悦竹.图像识别技术在铁路设备故障检测中的应用[J].信息技术,2013(7):147-149.
HUANG Feng, XU Yuezhu. Image recognition technology in the railway equipment fault detection[J].Information Technology, 2013(7): 147-149.
[6] 佟瑞鹏,陈策,崔鹏程,等.基于深度学习的施工安全泛场景数据获取方法[J].中国安全科学学报,2017,27(5):1-6.
TONG Ruipeng, CHEN Ce, CUI Pengcheng, et al. Deep learning method for processing pan-scene data on construction safety[J]. China Safety Science Journal, 2017, 27(5): 1-6.
[7] KRIZHEVSKY A. Convolutional deep belief networks on CIFAR-10[J]. Communications of the Acm, 2010, 54(10):95-103.
[8] BURGES C J C. From ranknet to lambdarank to lambdamart: an overview[J]. Learning, 2010, 11: 23-41.
[9] YAN J, XU N Y, CAI X F, et al. An FPGA-based accelerator for LambdaRank in Web search engines[J]. Acm Transactions on Reconfigurable Technology & Systems, 2011, 4(3): 2 658-2 674.
[10] 张瑜,罗可.基于OC-SVM的大型数据集分类方法[J].计算机工程与应用,2011,47(4):131-133.
ZHANG Yu, LUO Ke. OC-SVM-based classification for large-scale data sets[J]. Computer Engineering & Applications, 2011,47(4): 131-133.
[11] GRAY C, BEATTIE M, BELAY H, et al. Personalized online search for fashion products[C]. Systems and Information Engineering Design Symposium (SIEDS), 2015. IEEE, 2015:91-96.

[1]	张萌, 韩豫, 刘泽锋. 深度学习下建筑工人高空安全防护装备检测方法[J]. 中国安全科学学报, 2022, 32(5): 140-146.
[2]	郭玮, 袁宏永, 薛明, 魏平岩. SAR影像洪水淹没范围深度学习提取方法[J]. 中国安全科学学报, 2022, 32(4): 177-184.
[3]	古莹奎, 刘平, 林忠海, 邱光琦. 基于Kurtogram与DSCN的滚动轴承故障诊断方法[J]. 中国安全科学学报, 2021, 31(6): 99-105.
[4]	宋韬讲师, 夏正洪, 万健, 朱新平, 李阳助理研究员. 高原通导设施覆盖范围计算方法及可视化实现[J]. 中国安全科学学报, 2021, 31(4): 111-118.
[5]	任飞, 史聪灵, 李建, 车洪磊, 胥旋. 半地下有轨电车车站火灾自然排烟模式研究[J]. 中国安全科学学报, 2021, 31(3): 156-161.
[6]	段照斌, 杜海龙, 张鹏. 基于QAR2Vec模型的QAR数据特征提取[J]. 中国安全科学学报, 2021, 31(1): 145-152.
[7]	龙时丹. 面向高速铁路的全自动运行系统应用[J]. 中国安全科学学报, 2020, 30(S1): 92-97.
[8]	张帆, 步兵, 赵骏逸. 列车运行控制系统信息安全风险评估方法[J]. 中国安全科学学报, 2020, 30(S1): 172-178.
[9]	吕伟, 秦霜霜, 穆治国, 汪京辉, 姜雅娟, 李加伟. CRH380AL型高铁列车应急疏散及影响因素模拟研究[J]. 中国安全科学学报, 2020, 30(8): 164-170.
[10]	赵欣然, 张琪, 王卫东, 徐志强. 可燃性粉尘云的图像检测方法[J]. 中国安全科学学报, 2020, 30(4): 8-13.
[11]	文超, 李忠灿, 黄平, 田锐, 牟玮玮, 李力. 数据驱动的列车晚点传播研究^*[J]. 中国安全科学学报, 2019, 29(S2): 1-9.
[12]	汤轶雄, 徐传玲, 文超, 李忠灿, 宋邵杰. 高铁故障晚点时间预测的支持向量回归模型^*[J]. 中国安全科学学报, 2019, 29(S2): 18-23.
[13]	李力, 闫迪, 田锐, 王全泉. 基于晚点恢复的高速铁路行车调度质量评价^*[J]. 中国安全科学学报, 2019, 29(S2): 118-125.
[14]	余敬芝, 燕飞, 牛儒. 基于Petri网的形式化安全分析方法^*[J]. 中国安全科学学报, 2019, 29(S2): 138-143.
[15]	陈雪倩, 步兵. 基于网络流量和数据包的CBTC入侵检测系统^*[J]. 中国安全科学学报, 2019, 29(S2): 154-160.

基于深度学习算法的铁路列车运行安全检测^*

Study on safety inspection of railway train operation based on deep learning algorithm

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