Classification of Metro Facilities with Deep Neural Networks

He, Deqiang; Jiang, Zhou; Chen, Jiyong; Liu, Jian-Ren; Miao, Jianyin; Shah, Abid Ali

doi:10.1155/2019/6782803

Cited by 9 publications

(17 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The multilayer network structure can be used to continuously extract the characteristics of a sample dataset to achieve noncontact recognition of foreign objects. Reference [15] improved the deep convolutional neural network (CNN) to construct a subway operation detection network. Besides, it used transfer learning technology to train facility images in subway tunnels to improve the performance of obstacle model detection.…”

Section: Related Workmentioning

confidence: 99%

Subway Obstacle Perception and Identification Method Based on Cloud Edge Collaboration

Li¹,

Yan²,

Liu

et al. 2021

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

The traditional analysis method of train obstacle uses isomorphic sensors to obtain the state information and completes detection and identification analysis at the remote end of a network. A single data sample and more processing links will reduce the accuracy and speed analysis for subway encountering obstacles. To solve this problem, this paper proposes a subway obstacle perception and identification method based on cloud edge cooperation. The subway monitoring cloud platform realizes the training and construction of a detection model, and the network edge side completes the situation awareness of track state and real-time action when the train encounters obstacles. Firstly, the railroad track position is detected by cameras, and subway running track is identified by Mask RCNN algorithm to determine the detection area of obstacles in the process of subway train running. At the edge of network, the feature-level fusion of data collected by sensor cluster is carried out to provide reliable data support for detection work. Then, based on the DeepSort and YOLOv3 network models, the subway obstacle detection model is constructed on the subway monitoring cloud platform. Moreover, a trained model is distributed to the network edge side, so as to realize the fast and efficient perception and action of obstacles. Finally, the simulation verification is implemented based on actual collected datasets. Experimental results show that the proposed method has good detection accuracy and efficiency, which maintains 98.9% and 1.43 s for obstacle detection accuracy and recognition time in complex scenes.

show abstract

Section: Related Workmentioning

confidence: 99%

Subway Obstacle Perception and Identification Method Based on Cloud Edge Collaboration

Li¹,

Yan²,

Liu

et al. 2021

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…Therefore, it is necessary to solve the transitive closure t(C) of the fuzzy matrix. Starting from the transfer matrix C, 24 k…”

Section: Fuzzy Clustering Analysis Of Fault Datamentioning

confidence: 99%

“…Next, the weights of the characteristic attributes of the faults of the 8 types of selected fault modes are obtained. The weight of the characteristic attributes of the faults can be obtained by formula (24)…”

Section: Characteristic Attributes and Information Sequence Of Faultsmentioning

confidence: 99%

See 1 more Smart Citation

Preventive Maintenance Strategy for Train Doors Based on the Competitive Weibull Theory

he¹,

Zhang²,

Yan-jun³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

In view of the problems of over-maintenance and under-maintenance in the current urban rail transit maintenance strategy and the reliability of single processing of fault data, which is often inconsistent with the actual situation, an incomplete preventive maintenance strategy based on the competitive Weibull model is proposed in this paper. To make the fault mechanism processing method for urban rail vehicles more accurate, fault feature attributes and fault information sequences are introduced to classify fault data. Fuzzy cluster analysis of vehicle fault data can be performed using the formula of the competitive Weibull model, and parameter estimation of the reliability model can be performed by combining it with the graph parameter estimation method. In addition, the fault rate increase factor and service age reduction factor are introduced into the maintenance strategy, and the optimal preventive maintenance cycle and maintenance times are obtained by combining maintenance and replacement according to reliability. A quantum-genetic intelligent algorithm is used to optimize the model-solving process. Finally, the maintenance of urban rail transit train doors is taken as an example. The results of this study show that compared with the traditional maintenance strategy, the reliability of the proposed maintenance strategy is closer to the actual situation. At the same time, the proposed maintenance strategy can effectively reduce the number of parked vehicles, reduce maintenance costs, and ensure the safety of train operation, maintenance economy and performance of tasks.

show abstract

“…With the development of machine learning, deep learning technology has been widely used in railways. He et al [9] used a deep neural network model devised by Google's InceptionV3 to classify metro facilities. Wei et al [10] applied a method based on deep learning to detect railway track fasteners.…”

Section: Related Workmentioning

confidence: 99%

Detection of Foreign Matter on High-Speed Train Underbody Based on Deep Learning

Yao

Jiang

et al. 2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

When a high-speed train is running, it is easy for foreign objects like rail-side plastic bags to enter bottom bogies, cables and equipment gaps, which affects the safety of driving. At present, the detection accuracy of such foreign objects is low. To solve this problem, the present study used the latest deep learning based object detection networks, such as SSD and Faster R-CNN, which combined with different feature extractors to build a detection network. Through data augmentation of a sample, the number of sizes was expanded. By the idea of transfer learning, the COCO dataset was used as the source data to transfer features to the new detection network and then retrained to build a train bottom plastic bag detection network based on deep learning. It was shown that through data augmentation, the precision of each combination significantly improved. Moreover, the results using combination of SSD and MobileNet obtained the fastest detection speed with an average time of 41 ms and a precision rate of 81.3%. The accuracy using combination of SSD + Inception V2 combination reached to 89.7%, with an average time of 53 ms. INDEX TERMS Deep learning, foreign matter on underbody, transfer learning, data augmentation, highspeed train.

show abstract

Classification of Metro Facilities with Deep Neural Networks

Cited by 9 publications

References 16 publications

Subway Obstacle Perception and Identification Method Based on Cloud Edge Collaboration

Subway Obstacle Perception and Identification Method Based on Cloud Edge Collaboration

Preventive Maintenance Strategy for Train Doors Based on the Competitive Weibull Theory

Detection of Foreign Matter on High-Speed Train Underbody Based on Deep Learning

Contact Info

Product

Resources

About