A YOLOv3-based Deep Learning Application Research for Condition Monitoring of Rail Thermite Welded Joints

Liu, Yang; Sun, Xin; Pang, J.H.L.

doi:10.1145/3388818.3388827

Cited by 12 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the same reference may appear under different columns. [38], [39], [40], [35] Rails' Heads [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62] [63] † , [64] Fasteners/ Fastening Systems [48], [65], [66], [67], [68], [69], [70], [71] [72], [55] Welded Joints [21] † Sleepers [36] The † mark indicates data obtained from specimens only (laboratory tests)…”

Section: Rail Trackmentioning

confidence: 99%

“…Two papers deal with thermite welded joints detection and concrete and sleeper cracks width estimation. Reference [21] proposes a YOLOv3-based object detector for the welded joints detection task, trained and tested on images collected in laboratory thermite welded specimen. Reference [36] proposes three Semantic Segmentation models, based on the SegNet model, to predict the width of cracks in concrete and sleepers.…”

Section: Welded Joints and Sleepersmentioning

confidence: 99%

“…• in two cases, the authors "simply" focused on items classification or detection, without looking at defects. In particular, [72] focused on fasteners type classification, while, [21] focused on welded joints detection; • in two cases, tests were performed only on data coming from laboratory experiments (cf. [21], [63]); • some studies use geolocation to detect the location of faulty components on the rail tracks (e.g., [44], [69], [65]).…”

Section: E Findings On Rail Trackmentioning

confidence: 99%

See 2 more Smart Citations

A Survey on Audio-Video Based Defect Detection Through Deep Learning in Railway Maintenance

et al. 2022

View full text Add to dashboard Cite

Within Artificial Intelligence, Deep Learning (DL) represents a paradigm that has been showing unprecedented performance in image and audio processing by supporting or even replacing humans in defect and anomaly detection. The Railway sector is expected to benefit from DL applications, especially in predictive maintenance applications, where smart audio and video sensors can be leveraged yet kept distinct from safety-critical functions. Such separation is crucial, as it allows for improving system dependability with no impact on its safety certification. This is further supported by the development of DL in other transportation domains, such as automotive and avionics, opening for knowledge transfer opportunities and highlighting the potential of such a paradigm in railways. In order to summarize the recent state-of-the-art while inquiring about future opportunities, this paper reviews DL approaches for the analysis of data generated by acoustic and visual sensors in railway maintenance applications that have been published until August 31st, 2021. In this paper, the current state of the research is investigated and evaluated using a structured and systematic method, in order to highlight promising approaches and successful applications, as well as to identify available datasets, current limitations, open issues, challenges, and recommendations about future research directions.

show abstract

Section: Rail Trackmentioning

confidence: 99%

Section: Welded Joints and Sleepersmentioning

confidence: 99%

Section: E Findings On Rail Trackmentioning

confidence: 99%

See 1 more Smart Citation

A Survey on Audio-Video Based Defect Detection Through Deep Learning in Railway Maintenance

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Another trend in the application of image and video-processing methods is related to deep-learning computer vision algorithms such as YOLO3 and others [ 71 , 72 , 73 , 74 , 75 ]. They provide the object recognition, that in case of the sleeper support diagnostics could automatically bring the necessary information about the loading position, loading type, image quality improvement, etc.…”

Section: In Situ Measurements Of Rail Deflection In Void Zonementioning

confidence: 99%

Identification of Sleeper Support Conditions Using Mechanical Model Supported Data-Driven Approach

Sysyn

Przybyłowicz

Nabochenko

et al. 2021

Sensors

View full text Add to dashboard Cite

The ballasted track superstructure is characterized by a relative quick deterioration of track geometry due to ballast settlements and the accumulation of sleeper voids. The track zones with the sleeper voids differ from the geometrical irregularities with increased dynamic loading, high vibration, and unfavorable ballast-bed and sleeper contact conditions. This causes the accelerated growth of the inhomogeneous settlements, resulting in maintenance-expensive local instabilities that influence transportation reliability and availability. The recent identification and evaluation of the sleeper support conditions using track-side and on-board monitoring methods can help planning prevention activities to avoid or delay the development of local instabilities such as ballast breakdown, white spots, subgrade defects, etc. The paper presents theoretical and experimental studies that are directed at the development of the methods for sleeper support identification. The distinctive features of the dynamic behavior in the void zone compared to the equivalent geometrical irregularity are identified by numeric simulation using a three-beam dynamic model, taking into account superstructure and rolling stock dynamic interaction. The spectral features in time domain in scalograms and scattergrams are analyzed. Additionally, the theoretical research enabled to determine the similarities and differences of the dynamic interaction from the viewpoint of track-side and on-board measurements. The method of experimental investigation is presented by multipoint track-side measurements of rail-dynamic displacements using high-speed video records and digital imaging correlation (DIC) methods. The method is used to collect the statistical information from different-extent voided zones and the corresponding reference zones without voids. The applied machine learning methods enable the exact recent void identification using the wavelet scattering feature extraction from track-side measurements. A case study of the method application for an on-board measurement shows the moderate results of the recent void identification as well as the potential ways of its improvement.

show abstract

“…Raw data type: it is observed that 70% of studies used image-type raw data for the deep learning models. Nevertheless, acoustic emission signals [65,71,100,103,108] , defectogram [96,109] , speed accelerations [98] , concatenated vector of curve and numbers [101] , current signal [89] , maintenance records [80,99] , synthetic data from generative model [63] , time-frequency measurement data [82] , time-series [60] , geometry data [87] , and vibration signal [119] could all be possible input data sources as well. Purpose of study: it is observed that detection, classification, and/or localizing rail surface defects including various components (rail, insulator, valves, fasteners, switches, track intrusions, etc.)…”

Section: Review Of Rail Track Condition Monitoring With Deep Learningmentioning

confidence: 99%

Rail track condition monitoring: a review on deep learning approaches

Ji¹,

Woo²,

Wong³

et al. 2021

View full text Add to dashboard Cite

Rail track is a critical component of rail systems. Accidents or interruptions caused by rail track anomalies usually possess severe outcomes. Therefore, rail track condition monitoring is an important task. Over the past decade, deep learning techniques have been rapidly developed and deployed. In the paper, we review the existing literature on applying deep learning to rail track condition monitoring. Potential challenges and opportunities are discussed for the research community to decide on possible directions. Two application cases are presented to illustrate the implementation of deep learning to rail track condition monitoring in practice before we conclude the paper.

show abstract

A YOLOv3-based Deep Learning Application Research for Condition Monitoring of Rail Thermite Welded Joints

Cited by 12 publications

References 10 publications

A Survey on Audio-Video Based Defect Detection Through Deep Learning in Railway Maintenance

A Survey on Audio-Video Based Defect Detection Through Deep Learning in Railway Maintenance

Identification of Sleeper Support Conditions Using Mechanical Model Supported Data-Driven Approach

Rail track condition monitoring: a review on deep learning approaches

Contact Info

Product

Resources

About