Bionic Vision-Based Pantograph–Catenary Contact Point Detection Study in China High-Speed Railway

The pantograph–catenary system of high-speed electric multiple units (EMU) is the only way to get electric power for high-speed trains. Over the past several years, video frames taken from traditional cameras have been used to analyze pantograph–catenary arcs. A traditional camera is built on mimicking human vision; thus, the natural phenomenon that an object will appear smaller if it is far from an observer but will become larger as it moves toward the observer will be reflected in the pictures that are taken. This is analogous to an implicit depth. Based on this observation, a bionic vision-based algorithm that utilizes the implicit depth is proposed in the present work to extract the touch point between the contact wire and pantograph slide under interference from the messenger wire. Experiments indicate that the proposed algorithm works quite well, with only marginal errors occurring, thus providing a strong base for future research activities.

[1]  Yu Guo-wang Research on the Algorithm to Measure the Pantographic Slipper Abrasion , 2010 .

[2]  B. Hulin,et al.  Concepts for Day-Night Stereo Obstacle Detection in the Pantograph Gauge , 2007, 2007 5th IEEE International Conference on Industrial Informatics.

[3]  Zhigang Liu,et al.  An Extended Habedank’s Equation-Based EMTP Model of Pantograph Arcing Considering Pantograph-Catenary Interactions and Train Speeds , 2016, IEEE Transactions on Power Delivery.

[4]  Mehmet Karaköse,et al.  Image processing and model based arc detection in pantograph catenary systems , 2014, 2014 22nd Signal Processing and Communications Applications Conference (SIU).

[5]  Li Ma,et al.  Edge Detection on Pantograph Slide Image , 2009, 2009 2nd International Congress on Image and Signal Processing.

[6]  Jianxiong Xiao,et al.  SUN RGB-D: A RGB-D scene understanding benchmark suite , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Dieter Fox,et al.  RGB-D mapping: Using Kinect-style depth cameras for dense 3D modeling of indoor environments , 2012, Int. J. Robotics Res..

[8]  Yang Hong-mei,et al.  Overview of Non-contact Image Detection Technology for Pantograph-catenary Monitoring , 2013 .