Prediction of Rail Contact Fatigue on Crossings Using Image Processing and Machine Learning Methods

Abstract

In this paper, an application of computer vision and machine learning algorithms for common crossing frog diagnostics is presented. The rolling surface fatigue of frogs along the crossing lifecycle is analysed. The research is based on information from high-resolution optical images of the frog rolling surface and images from magnetic particle inspection. Image processing methods are used to pre-process the images and to detect the feature set that corresponds to objects similar to surface cracks. Machine learning methods are used for the analysis of crack images from the beginning to the end of the crossing lifecycle. Statistically significant crack features and their combinations that depict the surface fatigue state are found. The research result consists of the early prediction of rail contact fatigue.

Keywords

Railway turnout / Common crossing / Image processing / Rolling contact fatigue / Machine learning / Feature detection and selection

References

[1] Fendrich L, Fengler W. Handbuch Eisenbahninfrastruktur (Field manual Railway Infrastructure), 2013, Berlin: Springer

[2] Zoll A, Gerber U, Fengler W (2016) Das Messsystem ESAH-M (The measuring system ESAH-M). EI-Eisenbahningenieur Kalender 1:49–62. ISSN: 0934-5930

[3] Gerber U, Zoll A, Fengler W (2013) Fahrzeugbasierte Beurteilung des Herzstückverschleißes (Vehicle-based assessment of wear on common crossings). EI-Eisenbahningenieur 05:26–30. ISSN: 0013-2810

[4] Sysyn M, Kovalchuk V, Jiang D. Performance study of the inertial monitoring method for railway turnouts. Int J Rail Transp, 2019, 7(2): 103-116

[5] Gerber U, Zoll A, Fengler W (2015) Verschleiß und Fahrflächenermüdung an Weichen mit starrer Herzstückspitze (Wear and Rolling Contact Fatigue on common crossings of railway turnouts). ETR Eisenbahntechnische Rundschau 01:36–41. ISSN: 0013-2845

[6] Kovalchuk V, Sysyn M, Hnativ Y, Bal O, Parneta B, Pentsak A. Development of a promising system for diagnosing the frogs of railroad switches using the transverse profile measurement method. East Eur J Enterp Technol, 2018, 92(2): 33-42

[7] http://www.graw.com/downloadprod/product-leaflet-scorpion.pdf

[8] http://www.witt-solutions.de/assets/Witt_Broschuere_Bahn_a.pdf

[9] Kireev A. Increase of informativity of two-frequency method of defectometry at ultrasonic control of details and units of railway rolling stock. Vestn Railw Res Inst, 2018, 77(3): 182-187

[10] Sein S, Matos JC, Idnurm J. Statistical analysis of reinforced concrete bridges in Estonia. Baltic J Road Bridge Eng, 2017, 12(4): 225-233

[11] Soghomonyan V, Chibukhchyan G, Chibukhchyan H. Diagnostics of fatigue fractures of building structures elements. Mag Civ Eng, 2018, 4: 195-203

[12] Sysyn M, Nabochenko O, Kovalchuk V, Gerber U. Evaluation of railway ballast layer consolidation after maintenance works. Acta Polytech, 2019, 59(1): 77-87

[13] Storcz T, Ercsey Z, Várady G. Histogram based segmentation of shadowed leaf images. Pollack Period, 2018, 13(1): 22-32

[14] Hůlková M, Pavelka K, Matoušková E. Automatic classification of point clouds for highway documentation. Acta Polytech, 2018, 58(3): 165-170

[15] Shi Y, Cui L, Qi Z, Meng F, Chen Z. Automatic road crack detection using random structured forests. IEEE Trans Intell Transp Syst, 2016, 17(12): 3434-3445

[16] Bolkeny I, Fuvesi V. Ai based predictive detection system. Pollack Period, 2018, 13(2): 137-146

[17] Santur Y, Karaköse M, Akin E (2017) A new rail inspection method based on deep learning using laser cameras. In: IDAP 2017—international artificial intelligence and data processing symposium, 8090245. https://doi.org/10.1109/IDAP.2017.8090245

[18] Hajizadeh S, Núñez A, Tax D. Semi-supervised rail defect detection from imbalanced image Data. IFAC-PapersOnLine, 2016, 49(3): 78-83

[19] Wang L, Hang Y, Luo S, Luo X, Jiang X (2011) Deblurring gaussian-blur images: a preprocessing for rail head surface defect detection. In: Proceedings of 2011 IEEE international conference on service operations, logistics and informatics, vol 59, pp 451–456. https://doi.org/10.1109/SOLI.2011.5986603

[20] Sysyn M, Gruen D, Gerber U, Nabochenko O, Kovalchuk V. Turnout monitoring with vehicle based inertial measurements of operational trains: a machine learning approach. Commun Sci Lett Univ Zilina, 2019, 21(1): 42-48.

[21] Sysyn M, Gerber U, Gruen D, Nabochenko O, Kovalchuk V. Modelling and vehicle based measurements of ballast settlements under the common crossing. Eur Transp Int J Transp Econ Eng Law, 2019, 71: 1-25.

[22] Oge M. Practical image and video processing using MATLAB, 2011, New York: Wiley-IEEE Press 299-335.

[23] Kong X, Hu C, Duan Z. Principal component analysis networks and algorithms, 2017, Singapore: Springer

[24] Bramer M. Principles of data mining, 2016 3 London: Springer 1-526

[25] James G, Witten D, Hastie T, Tibshirani R. An introduction to statistical learning with applications in R, 2013, New York: Springer 373-385