Abstract
The article presents a vision system for detecting elements of railway track. Four types of fasteners, wooden and concrete sleepers, rails, and turnouts can be recognized by our system. In addition, it is possible to determine the degree of sleeper ballast coverage. Our system is also able to work when the track is moderately covered by snow. We used a Fully Convolutional Neural Network with 8 times upsampling (FCN-8) to detect railway track elements. In order to speed up training and improve performance of the model, a pre-trained deep convolutional neural network developed by Oxford’s Visual Geometry Group (VGG16) was used as a framework for our system. We also verified the invariance of our system to changes in brightness. To do this, we artificially varied the brightness of images. We performed two types of tests. In the first test, we changed the brightness by a constant value for the whole analyzed image. In the second test, we changed the brightness according to a predefined distribution corresponding to Gaussian function.
Highlights
A steady increase both in tonnage and people carried by trains requires that an emphasis be put on the state of the railway track infrastructure
We used a pre-trained VGG-16 network as a framework to develop our visual system based on Fully Convolutional Network with 8 times upsampling (FCN-8)
The visual system based on Fully Convolutional Network (FCN)-8 network presented in this paper allows for the detection the following elements of railway track: fasteners, wooden and concrete sleepers, rails, and turnouts
Summary
A steady increase both in tonnage and people carried by trains requires that an emphasis be put on the state of the railway track infrastructure. Surface diagnostics of individual elements, including the rails, makes it possible to determine their technical condition For this purpose, different methods are used, one of them being the vision method [5,6,7]. Recent advances in CMOS imaging technology and Graphics Processing Units have resulted in commercial line-scan cameras and highly efficient graphics servers that can be used as elements of a visual system to inspect railway track elements. The authors use a Fully Convolutional Network with 8 times upsampling (FCN-8) to segment the image of railway track into regions containing ballast, sleepers, fasteners, rail, and turn-out. Evaluation of FCN-8 network performance at different levels of light intensity of railway track images; and. Evaluation of FCN-8 network performance at different distributions of light intensity of railway track images
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
