Automatic inspection of railway carbon strips based on multi-modal visual information

Electric rail vehicles are driven by the current collected from the electrical lines of the railway catenary system. For this purpose, trains are equipped with a current collector (the pantograph) that comes in contact with the wires by means of a pair of carbon strips. The sliding movement along the wires subjects the carbon strips to wear and damage so that frequent inspection is essential to ensure the train and catenary safety. In this paper we describe an automatic visual inspection system, made of a 3D laser scanner and a 2D high resolution camera, which is able to automatically determine the health state of the railway contact strips thanks to a sophisticated data processing tool. The system collects color and geometrics information on the carbon strips and performs several automated assessments both on the 3D and 2D data. For each test, the system provides an index and decides between three different levels of wear (low, medium, high) to support the human operator in inspection and maintenance decisions. Experimental results reveal the effectiveness of the system, comparing the human judgment with the automated analysis.

[1]  Carlo Alberto Avizzano,et al.  Automatic 2D-3D vision based assessment of the attitude of a train pantograph , 2016, 2016 IEEE International Smart Cities Conference (ISC2).

[2]  Yacov Hel-Or,et al.  Multiple histogram matching , 2013, 2013 IEEE International Conference on Image Processing.

[3]  W. Kortüm,et al.  Pantograph/Catenary Dynamics and Control , 1997 .

[4]  Linda G. Shapiro,et al.  Image Segmentation Techniques , 1984, Other Conferences.

[5]  Mehmet Karaköse,et al.  A New Experimental Approach Using Image Processing-Based Tracking for an Efficient Fault Diagnosis in Pantograph–Catenary Systems , 2017, IEEE Transactions on Industrial Informatics.

[6]  S. Cagnoni,et al.  PAVISYS : A computer vision system for the inspection of locomotive pantographs , 2011 .

[7]  Leonard G. C. Hamey,et al.  Pancam: In-Service Inspection of Locomotive Pantographs , 2007, 9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications (DICTA 2007).

[8]  Leszek Jarzebowicz,et al.  Wavelet transform-based approach to defect identification in railway carbon contact strips. , 2015 .

[9]  Hanan Samet,et al.  Efficient Component Labeling of Images of Arbitrary Dimension Represented by Linear Bintrees , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Patrick F. Dunn,et al.  Measurement and Data Analysis for Engineering and Science , 2017 .

[11]  Slawomir Judek,et al.  3D machine vision system for inspection of contact strips in railway vehicle current collectors , 2014, 2014 International Conference on Applied Electronics.

[12]  Edward R. Dougherty,et al.  An introduction to morphological image processing , 1992 .

[13]  J. Canny.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Y. Meyer.  Wavelets and Operators , 1993 .

[15]  Stefan Östlund,et al.  Condition monitoring of pantograph contact strip , 2008 .

[16]  I. Aydin,et al.  A new contactless fault diagnosis approach for pantograph-catenary system , 2012, Proceedings of 15th International Conference MECHATRONIKA.

[17]  Slawomir Judek,et al.  Algorithm for automatic wear estimation of railway contact strips based on 3D scanning results , 2014, 2014 International Conference and Exposition on Electrical and Power Engineering (EPE).