Classification of Metro Facilities with Deep Neural Networks

Metro barrier-detection has been one of the most popular research fields. How to detect obstacles quickly and accurately during metro operation is the key issue in the study of automatic train operation. Intelligent monitoring systems based on computer vision not only complete safeguarding tasks eff...

Full description

Saved in:
Bibliographic Details
Main Authors: Deqiang He, Zhou Jiang, Jiyong Chen, Jianren Liu, Jian Miao, Abid Shah
Format: Article
Language:English
Published: Wiley 2019-01-01
Series:Journal of Advanced Transportation
Online Access:http://dx.doi.org/10.1155/2019/6782803
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850219037014884352
author Deqiang He
Zhou Jiang
Jiyong Chen
Jianren Liu
Jian Miao
Abid Shah
author_facet Deqiang He
Zhou Jiang
Jiyong Chen
Jianren Liu
Jian Miao
Abid Shah
author_sort Deqiang He
collection DOAJ
description Metro barrier-detection has been one of the most popular research fields. How to detect obstacles quickly and accurately during metro operation is the key issue in the study of automatic train operation. Intelligent monitoring systems based on computer vision not only complete safeguarding tasks efficiently but also save a great deal of human labor. Deep convolutional neural networks (DCNNs) are the most state-of-the-art technology in computer vision tasks. In this paper, we evaluated the effectiveness in classifying the common facility images in metro tunnels based on Google’s Inception V3 DCNN. The model requires fewer computational resources. The number of parameters and the computational complexity are much smaller than similar DCNNs. We changed its architecture (the last softmax layer and the auxiliary classifier) and used transfer learning technology to retrain the common facility images in the metro tunnel. We use mean average precision (mAP) as the metric for performance evaluation. The results indicate that our recognition model achieved 90.81% mAP. Compared with the existing method, this method is a considerable improvement.
format Article
id doaj-art-7917ca3a2d914bfd9b6c174a45da7b3e
institution OA Journals
issn 0197-6729
2042-3195
language English
publishDate 2019-01-01
publisher Wiley
record_format Article
series Journal of Advanced Transportation
spelling doaj-art-7917ca3a2d914bfd9b6c174a45da7b3e2025-08-20T02:07:31ZengWileyJournal of Advanced Transportation0197-67292042-31952019-01-01201910.1155/2019/67828036782803Classification of Metro Facilities with Deep Neural NetworksDeqiang He0Zhou Jiang1Jiyong Chen2Jianren Liu3Jian Miao4Abid Shah5Guangxi Key Laboratory of Manufacturing System & Advanced Manufacturing Technology, College of Mechanical Engineering, Guangxi University, 530004, Nanning, ChinaGuangxi Key Laboratory of Manufacturing System & Advanced Manufacturing Technology, College of Mechanical Engineering, Guangxi University, 530004, Nanning, ChinaGuangxi Key Laboratory of Manufacturing System & Advanced Manufacturing Technology, College of Mechanical Engineering, Guangxi University, 530004, Nanning, ChinaGuangxi Key Laboratory of Manufacturing System & Advanced Manufacturing Technology, College of Mechanical Engineering, Guangxi University, 530004, Nanning, ChinaGuangxi Key Laboratory of Manufacturing System & Advanced Manufacturing Technology, College of Mechanical Engineering, Guangxi University, 530004, Nanning, ChinaGuangxi Key Laboratory of Manufacturing System & Advanced Manufacturing Technology, College of Mechanical Engineering, Guangxi University, 530004, Nanning, ChinaMetro barrier-detection has been one of the most popular research fields. How to detect obstacles quickly and accurately during metro operation is the key issue in the study of automatic train operation. Intelligent monitoring systems based on computer vision not only complete safeguarding tasks efficiently but also save a great deal of human labor. Deep convolutional neural networks (DCNNs) are the most state-of-the-art technology in computer vision tasks. In this paper, we evaluated the effectiveness in classifying the common facility images in metro tunnels based on Google’s Inception V3 DCNN. The model requires fewer computational resources. The number of parameters and the computational complexity are much smaller than similar DCNNs. We changed its architecture (the last softmax layer and the auxiliary classifier) and used transfer learning technology to retrain the common facility images in the metro tunnel. We use mean average precision (mAP) as the metric for performance evaluation. The results indicate that our recognition model achieved 90.81% mAP. Compared with the existing method, this method is a considerable improvement.http://dx.doi.org/10.1155/2019/6782803
spellingShingle Deqiang He
Zhou Jiang
Jiyong Chen
Jianren Liu
Jian Miao
Abid Shah
Classification of Metro Facilities with Deep Neural Networks
Journal of Advanced Transportation
title Classification of Metro Facilities with Deep Neural Networks
title_full Classification of Metro Facilities with Deep Neural Networks
title_fullStr Classification of Metro Facilities with Deep Neural Networks
title_full_unstemmed Classification of Metro Facilities with Deep Neural Networks
title_short Classification of Metro Facilities with Deep Neural Networks
title_sort classification of metro facilities with deep neural networks
url http://dx.doi.org/10.1155/2019/6782803
work_keys_str_mv AT deqianghe classificationofmetrofacilitieswithdeepneuralnetworks
AT zhoujiang classificationofmetrofacilitieswithdeepneuralnetworks
AT jiyongchen classificationofmetrofacilitieswithdeepneuralnetworks
AT jianrenliu classificationofmetrofacilitieswithdeepneuralnetworks
AT jianmiao classificationofmetrofacilitieswithdeepneuralnetworks
AT abidshah classificationofmetrofacilitieswithdeepneuralnetworks