Fault detection of railway freight cars mechanical components based on multi-feature fusion convolutional neural network

Tao, Ye; Zhang, Xi; Chen, Yongran; Zhou, Fuqiang

doi:10.1007/s13042-021-01274-z

Cited by 18 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A fast adaptive Markov random field (FAMRF) algorithm and an exact height function (EHF) [8] were used to detect faults in the braking system, which also compared the performance with a cascade detector based on local binary patterns (LBPs). Ye et al [9] proposed a network of multi-feature fusion to detect three typical faults. The detection accuracy reached 88.72% and had good robustness to complex noise environments.…”

Section: A Fault Detection Of Freight Train Imagesmentioning

confidence: 99%

Visual Fault Detection of Multiscale Key Components in Freight Trains

Zhang

Zhou

Pan

et al. 2023

IEEE Trans. Ind. Inf.

View full text Add to dashboard Cite

Fault detection for key components in the braking system of freight trains is critical for ensuring railway transportation safety. Despite the frequently employed methods based on deep learning, these fault detectors are extremely reliant on hardware resources and complex to implement. In addition, no train fault detectors consider the drop in accuracy induced by scale variation of fault parts. This paper proposes a lightweight anchor-free framework to solve the above problems. Specifically, to reduce the amount of computation and model size, we introduce a lightweight backbone and adopt an anchor-free method for localization and regression. To improve detection accuracy for multi-scale parts, we design a feature pyramid network to generate rectangular layers of different sizes to map parts with similar aspect ratios. Experiments on four fault datasets show that our framework achieves 98.44% accuracy while the model size is only 22.5 MB, outperforming state-of-the-art detectors.

show abstract

Section: A Fault Detection Of Freight Train Imagesmentioning

confidence: 99%

Visual Fault Detection of Multiscale Key Components in Freight Trains

Zhang

Zhou

Pan

et al. 2023

IEEE Trans. Ind. Inf.

View full text Add to dashboard Cite

show abstract

“…The efficacy of this method is assessed within the context of online script recognition, with the categorization outcomes compared against a previous method. The proposed approach generates an output vector for each distance classifier, making it suitable for statistically tailored classifiers across diverse pattern-matching contexts [16].…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%

Neural Network-Directed Detection and Localization of Faults in Railway Track Circuits: An Application of Dempster-Shafer Theory

Mhmood,

Nasret,

Noori

et al. 2023

ISI

View full text Add to dashboard Cite

Track circuits, integral to the security infrastructure of railway traffic systems, govern the operational status of train lines. The swift detection and rectification of faults within these circuits is critical to preserve the integrity and functionality of rail networks. In this study, an innovative approach, leveraging neural networks in tandem with Dempster-Shafer theory, is proposed for detecting and localizing faults in track circuits. The complexities of fault detection are deconstructed into more manageable, capacitor-specific pattern recognition challenges, with the resolutions amalgamated via Dempster-Shafer theory. Simulations demonstrate the efficacy of this method, yielding a detection accuracy exceeding 98% and a localization accuracy surpassing 93%. This marks a significant improvement over contemporary reference techniques, thereby setting a new benchmark in the domain of track circuit fault detection and localization.

show abstract

“…Ling et al [14] proposed an instance detection model based on hierarchical features, which can perform instance-level prediction of rough defect areas. Ye et al [9] proposed a multi-feature fusion network for simultaneous detection of three typical mechanical component failures, which has good robustness to complex environments. Pahwa et al [5] proposed a method to detect valves failure using image segmentation followed by neural network recognition.…”

Section: A Fault Detection Of Freight Train Imagesmentioning

confidence: 99%

A Lightweight NMS-free Framework for Real-time Visual Fault Detection System of Freight Trains

Sun,

Zhou,

Pan

et al. 2022

Preprint

View full text Add to dashboard Cite

Real-time vision-based system of fault detection (RVBS-FD) for freight trains is an essential part of ensuring railway transportation safety. Most existing vision-based methods still have high computational costs based on convolutional neural networks. The computational cost is mainly reflected in the backbone, neck, and post-processing, i.e., non-maximum suppression (NMS). In this paper, we propose a lightweight NMS-free framework to achieve real-time detection and high accuracy simultaneously. First, we use a lightweight backbone for feature extraction and design a fault detection pyramid to process features. This fault detection pyramid includes three novel individual modules using attention mechanism, bottleneck, and dilated convolution for feature enhancement and computation reduction. Instead of using NMS, we calculate different loss functions, including classification and location costs in the detection head, to further reduce computation. Experimental results show that our framework achieves over 83 frames per second speed with a smaller model size and higher accuracy than the state-of-the-art detectors. Meanwhile, the hardware resource requirements of our method are low during the training and testing process.

show abstract

Fault detection of railway freight cars mechanical components based on multi-feature fusion convolutional neural network

Cited by 18 publications

References 21 publications

Visual Fault Detection of Multiscale Key Components in Freight Trains

Visual Fault Detection of Multiscale Key Components in Freight Trains

Neural Network-Directed Detection and Localization of Faults in Railway Track Circuits: An Application of Dempster-Shafer Theory

A Lightweight NMS-free Framework for Real-time Visual Fault Detection System of Freight Trains

Contact Info

Product

Resources

About