SCueU-Net: Efficient Damage Detection Method for Railway Rail

Lu, Jun; Liang, Bo; Lei, Qujiang; Li, Xiuhao; Liu, Junhao; Ji, Liu; Xu, Jie

doi:10.1109/access.2020.3007603

Cited by 39 publications

(24 citation statements)

References 60 publications

(73 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The proposed system autonomously detected the bolts that secure the rail to the sleepers and monitored the rail condition at high speeds, up to 460 km/h. VDSs are widely used for rail surface defect detection [ 22 , 23 , 24 , 25 ]. There are studies where the detection of defects of the railway plugs and fasteners using VDSs is analysed [ 26 , 27 ].…”

Section: Related Workmentioning

confidence: 99%

Visual Measurement System for Wheel–Rail Lateral Position Evaluation

Skrickij

Šabanovič

Shi

et al. 2021

Sensors

View full text Add to dashboard Cite

Railway infrastructure must meet safety requirements concerning its construction and operation. Track geometry monitoring is one of the most important activities in maintaining the steady technical conditions of rail infrastructure. Commonly, it is performed using complex measurement equipment installed on track-recording coaches. Existing low-cost inertial sensor-based measurement systems provide reliable measurements of track geometry in vertical directions. However, solutions are needed for track geometry parameter measurement in the lateral direction. In this research, the authors developed a visual measurement system for track gauge evaluation. It involves the detection of measurement points and the visual measurement of the distance between them. The accuracy of the visual measurement system was evaluated in the laboratory and showed promising results. The initial field test was performed in the Vilnius railway station yard, driving at low velocity on the straight track section. The results show that the image point selection method developed for selecting the wheel and rail points to measure distance is stable enough for TG measurement. Recommendations for the further improvement of the developed system are presented.

show abstract

Section: Related Workmentioning

confidence: 99%

Visual Measurement System for Wheel–Rail Lateral Position Evaluation

Skrickij

Šabanovič

Shi

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…The non-conforming and unexpected patterns are called outliers or anomalies. Anomaly detection has been extensively used in a wide variety of applications such as cyber-intrusion detection [31], fraud detection [32], medical anomaly detection [33,34], industrial damage detection [35], hyperspectral image analysis [36], sensor networks [37], image processing [38], to cite just a few. Outlier detection is very popular in industrial applications [39][40][41][42][43] since it is critical to the efficient and secure operation of industrial equipment, integrated sensors, and the overall production process.…”

Section: Related Workmentioning

confidence: 99%

Online Learning of Oil Leak Anomalies in Wind Turbines with Block-Based Binary Reservoir

et al. 2021

View full text Add to dashboard Cite

The focus of this work is to design a deeply quantized anomaly detector of oil leaks that may happen at the junction between the wind turbine high-speed shaft and the external bracket of the power generator. We propose a block-based binary shallow echo state network (BBS-ESN) architecture belonging to the reservoir computing (RC) category and, as we believe, it also extends the extreme learning machines (ELM) domain. Furthermore, BBS-ESN performs binary block-based online training using fixed and minimal computational complexity to achieve low power consumption and deployability on an off-the-shelf micro-controller (MCU). This has been achieved through binarization of the images and 1-bit quantization of the network weights and activations. 3D rendering has been used to generate a novel publicly available dataset of photo-realistic images similar to those potentially acquired by image sensors on the field while monitoring the junction, without and with oil leaks. Extensive experimentation has been conducted using a STM32H743ZI2 MCU running at 480 MHz and the results achieved show an accurate identification of anomalies, with a reduced computational cost per image and memory occupancy. Based on the obtained results, we conclude that BBS-ESN is feasible on off-the-shelf 32 bit MCUs. Moreover, the solution is also scalable in the number of image cameras to be deployed and to achieve accurate and fast oil leak detections from different viewpoints.

show abstract

“…The improved method can obtain good performance, but it has disadvantages such as a slow detection speed and large detection model. Lu et al [20] proposed to apply the combined U-Net graph segmentation network and damage location method for damage detection of high-speed railways. This method can obtain a high detection accuracy but has the limitations of slow detection speed and large model volume.…”

Section: Introductionmentioning

confidence: 99%

A Study on Railway Surface Defects Detection Based on Machine Vision

Bai

Gao

Yang

et al. 2021

Entropy

View full text Add to dashboard Cite

The detection of rail surface defects is an important tool to ensure the safe operation of rail transit. Due to the complex diversity of track surface defect features and the small size of the defect area, it is difficult to obtain satisfying detection results by traditional machine vision methods. The existing deep learning-based methods have the problems of large model sizes, excessive parameters, low accuracy and slow speed. Therefore, this paper proposes a new method based on an improved YOLOv4 (You Only Look Once, YOLO) for railway surface defect detection. In this method, MobileNetv3 is used as the backbone network of YOLOv4 to extract image features, and at the same time, deep separable convolution is applied on the PANet layer in YOLOv4, which realizes the lightweight network and real-time detection of the railway surface. The test results show that, compared with YOLOv4, the study can reduce the amount of the parameters by 78.04%, speed up the detection by 10.36 frames per second and decrease the model volume by 78%. Compared with other methods, the proposed method can achieve a higher detection accuracy, making it suitable for the fast and accurate detection of railway surface defects.

show abstract

SCueU-Net: Efficient Damage Detection Method for Railway Rail

Cited by 39 publications

References 60 publications

Visual Measurement System for Wheel–Rail Lateral Position Evaluation

Visual Measurement System for Wheel–Rail Lateral Position Evaluation

Online Learning of Oil Leak Anomalies in Wind Turbines with Block-Based Binary Reservoir

A Study on Railway Surface Defects Detection Based on Machine Vision

Contact Info

Product

Resources

About