Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current Defectograms

To ensure traffic safety of railway transport, non-destructive test of rails is regularly carried out by using various approaches and methods, including eddy-current flaw detection methods. An automatic analysis of large data sets (defectograms) that come from the corresponding equipment is an actua...

Full description

Saved in:
Bibliographic Details
Main Authors: Egor V. Kuzmin, Oleg E. Gorbunov, Petr O. Plotnikov, Vadim A. Tyukin, Vladimir A. Bashkin
Format: Article
Language:English
Published: Yaroslavl State University 2020-09-01
Series:Моделирование и анализ информационных систем
Subjects:
Online Access:https://www.mais-journal.ru/jour/article/view/1351
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850025689162448896
author Egor V. Kuzmin
Oleg E. Gorbunov
Petr O. Plotnikov
Vadim A. Tyukin
Vladimir A. Bashkin
author_facet Egor V. Kuzmin
Oleg E. Gorbunov
Petr O. Plotnikov
Vadim A. Tyukin
Vladimir A. Bashkin
author_sort Egor V. Kuzmin
collection DOAJ
description To ensure traffic safety of railway transport, non-destructive test of rails is regularly carried out by using various approaches and methods, including eddy-current flaw detection methods. An automatic analysis of large data sets (defectograms) that come from the corresponding equipment is an actual problem. The analysis means a process of determining the presence of defective sections along with identifying structural elements of railway tracks in defectograms. This article is devoted to the problem of recognizing images of long structural elements of rails in eddy-current defectograms. Two classes of rail track structural elements are considered: 1) rolling stock axle counters, 2) rail crossings. Long marks that cannot be assigned to these two classes are conditionally considered as defects and are placed in a separate third class. For image recognition of structural elements in defectograms a convolutional neural network is applied. The neural network is implemented by using the open library TensorFlow. To this purpose each selected (picked out) area of a defectogram is converted into a graphic image in a grayscale with size of 30 x 140 points.
format Article
id doaj-art-e8ff5fe6e5a5431fa77b74b2f0a71422
institution DOAJ
issn 1818-1015
2313-5417
language English
publishDate 2020-09-01
publisher Yaroslavl State University
record_format Article
series Моделирование и анализ информационных систем
spelling doaj-art-e8ff5fe6e5a5431fa77b74b2f0a714222025-08-20T03:00:45ZengYaroslavl State UniversityМоделирование и анализ информационных систем1818-10152313-54172020-09-0127331632910.18255/1818-1015-2020-3-316-3291010Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current DefectogramsEgor V. Kuzmin0Oleg E. Gorbunov1Petr O. Plotnikov2Vadim A. Tyukin3Vladimir A. Bashkin4P.G. Demidov Yaroslavl State UniversityCenter of Innovative Programming, NDDLabCenter of Innovative Programming, NDDLabCenter of Innovative Programming, NDDLabP.G. Demidov Yaroslavl State UniversityTo ensure traffic safety of railway transport, non-destructive test of rails is regularly carried out by using various approaches and methods, including eddy-current flaw detection methods. An automatic analysis of large data sets (defectograms) that come from the corresponding equipment is an actual problem. The analysis means a process of determining the presence of defective sections along with identifying structural elements of railway tracks in defectograms. This article is devoted to the problem of recognizing images of long structural elements of rails in eddy-current defectograms. Two classes of rail track structural elements are considered: 1) rolling stock axle counters, 2) rail crossings. Long marks that cannot be assigned to these two classes are conditionally considered as defects and are placed in a separate third class. For image recognition of structural elements in defectograms a convolutional neural network is applied. The neural network is implemented by using the open library TensorFlow. To this purpose each selected (picked out) area of a defectogram is converted into a graphic image in a grayscale with size of 30 x 140 points.https://www.mais-journal.ru/jour/article/view/1351nondestructive testingeddy-current testingrail flaw detectionautomated analysis of defectogramsneural networks
spellingShingle Egor V. Kuzmin
Oleg E. Gorbunov
Petr O. Plotnikov
Vadim A. Tyukin
Vladimir A. Bashkin
Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current Defectograms
Моделирование и анализ информационных систем
nondestructive testing
eddy-current testing
rail flaw detection
automated analysis of defectograms
neural networks
title Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current Defectograms
title_full Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current Defectograms
title_fullStr Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current Defectograms
title_full_unstemmed Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current Defectograms
title_short Application of Convolutional Neural Networks for Recognizing Long Structural Elements of Rails in Eddy-Current Defectograms
title_sort application of convolutional neural networks for recognizing long structural elements of rails in eddy current defectograms
topic nondestructive testing
eddy-current testing
rail flaw detection
automated analysis of defectograms
neural networks
url https://www.mais-journal.ru/jour/article/view/1351
work_keys_str_mv AT egorvkuzmin applicationofconvolutionalneuralnetworksforrecognizinglongstructuralelementsofrailsineddycurrentdefectograms
AT olegegorbunov applicationofconvolutionalneuralnetworksforrecognizinglongstructuralelementsofrailsineddycurrentdefectograms
AT petroplotnikov applicationofconvolutionalneuralnetworksforrecognizinglongstructuralelementsofrailsineddycurrentdefectograms
AT vadimatyukin applicationofconvolutionalneuralnetworksforrecognizinglongstructuralelementsofrailsineddycurrentdefectograms
AT vladimirabashkin applicationofconvolutionalneuralnetworksforrecognizinglongstructuralelementsofrailsineddycurrentdefectograms