Train rail defect classification detection and its parameters learning method

Wu, Fupei; Li, Qinghua; Li, Shengping; Wu, Tao

doi:10.1016/j.measurement.2019.107246

Cited by 38 publications

(18 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Table 7 shows the different parameters and indicators, and the comparison between the existing methods and our system. As can be seen from the table, although the traditional mathematical modeling [29] was faster than the deep-learning-based mapping, its accuracy was not high because it could not adapt to most nonlinear distortion relations. If the distortion correction was carried out without the down-sampling operation, we can see that a large amount of processing time was lost, while the accuracy and SSIM exponential increase was not high.…”

Section: Resultsmentioning

confidence: 99%

An End-To-End Model for Pipe Crack Three-Dimensional Visualization Based on a Cascade Neural Network

Xia

Wang

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

With the continuous progress of machine vision technology, crack detection in pipelines has been greatly improved. For crack detection in deep holes, inner tubes, and other environments, it is not only necessary to detect the existence of cracks, but also to collect important information regarding the crack detection direction for further analysis. Because shooting with a frontal field of view causes the real side wall images to produce certain distortions, the detection and calibration of cracks requires a certain amount of professional technology and time. It usually takes a long time to collect the image to eliminate the distortion, and then to identify the crack and mark the direction according to the data line. Therefore, a simple and efficient end-to-end neural network model for crack recognition and three-dimensional visualization are proposed by using a cascade network and simple recognition technology in conjunction with inertial navigation equipment. In addition, we screen the crack data via pixel calibration and eliminate the ambiguous data to make the visualization more accurate. Experiments in pipelines and burrows show that the accuracy, performance, and efficiency of the proposed method reached a high level.

show abstract

Section: Resultsmentioning

confidence: 99%

An End-To-End Model for Pipe Crack Three-Dimensional Visualization Based on a Cascade Neural Network

Xia

Wang

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The significant issue is that complicated background noises pollute the defect images. Diverse hand-crafted processing approaches (e.g., [ 15 , 16 ]) have been developed to extract defect features, but are either sensitive to noises or dependent on expert experience. A coarse-to-fine model was established by Yu et al [ 17 ] for defect detection, but with several significant parameters determined by experience.…”

Section: Introductionmentioning

confidence: 99%

Wavelet Scattering and Neural Networks for Railhead Defect Identification

Yang

2021

Materials

View full text Add to dashboard Cite

Accurate and automatic railhead inspection is crucial for the operational safety of railway systems. Deep learning on visual images is effective in the automatic detection of railhead defects, but either intensive data requirements or ignoring defect sizes reduce its applicability. This paper developed a machine learning framework based on wavelet scattering networks (WSNs) and neural networks (NNs) for identifying railhead defects. WSNs are functionally equivalent to deep convolutional neural networks while containing no parameters, thus suitable for non-intensive datasets. NNs can restore location and size information. The publicly available rail surface discrete defects (RSDD) datasets were analyzed, including 67 Type-I railhead images acquired from express tracks and 128 Type-II images captured from ordinary/heavy haul tracks. The ultimate validation accuracy reached 99.80% and 99.44%, respectively. WSNs can extract implicit signal features, and the support vector machine classifier can improve the learning accuracy of NNs by over 6%. Three criteria, namely the precision, recall, and F-measure, were calculated for comparison with the literature. At the pixel level, the developed approach achieved three criteria of around 90%, outperforming former methods. At the defect level, the recall rates reached 100%, indicating all labeled defects were identified. The precision rates were around 75%, affected by the insignificant misidentified speckles (smaller than 20 pixels). Nonetheless, the developed learning framework was effective in identifying railhead defects.

show abstract

“…Rail defect research is pivotal for railway companies [1]. Therefore, they have put plenty of effort into rail defect detection and related maintenance [2][3][4][5][6][7][8][9][10][11]. This research presents a new type of data-driven method for rail defects.…”

Section: Introductionmentioning

confidence: 99%

“…The research is limited to the surface defects, and it is not related to manufacturing technologies of welds and the materials of welds [27]. Other researchers have also not claimed any success or viable solutions to the problems we are working on [2][3][4][5][6][7][8][9][10][11][21][22][28][29][30]. We first extract all the key features related to the problems and first utilize data mining approaches for weld defect prediction problems.…”

Section: Introductionmentioning

confidence: 99%

Rail Weld Defect Prediction and Related Condition-Based Maintenance

Yao

Jia²,

Tao

2020

IEEE Access

View full text Add to dashboard Cite

Rail weld defects are major threats to railroad transportation. Enormous resources have been required for related maintenance. This paper presents a creative solution to predict weld defects and to classify railroads into different conditions based on the predictions. The results are based on features extracted from manufacturing technologies of welds, from related materials and from influential factors in the environments. Features such as marks for welding engineers are defined. Maintenance can be selectively implemented based on the predicted conditions. Safety is the foundation of the railroad business, and a very strict safety requirement is utilized as one of the main constraints in this research. Additionally, 11 key risk factors leading to rail defects and their risk levels are identified. Extreme learning machine (ELM), random forest, logistic regression, principal component analysis (PCA), support vector machine (SVM) and other data science approaches are utilized. The evaluation results show that the related rail maintenance workload can decrease significantly under high safety standards. Labor costs of weld inspection will be reduced substantially because of the decreased workload for the sections predicted to not have any defects with a 100% recall rate (approximately 30% of the total sections), contributing to a massive cost reduction. Consequently, rail companies are expected to achieve enhanced management and operation.

show abstract

Train rail defect classification detection and its parameters learning method

Cited by 38 publications

References 17 publications

An End-To-End Model for Pipe Crack Three-Dimensional Visualization Based on a Cascade Neural Network

An End-To-End Model for Pipe Crack Three-Dimensional Visualization Based on a Cascade Neural Network

Wavelet Scattering and Neural Networks for Railhead Defect Identification

Rail Weld Defect Prediction and Related Condition-Based Maintenance

Contact Info

Product

Resources

About