The Simulation Of An Installation Position Of Wayside Signals Using Video Sequences From The Train Cab
Price
Free (open access)
Transaction
Volume
155
Pages
13
Page Range
97 - 109
Published
2015
Size
1,080 kb
Paper DOI
10.2495/CRS140091
Copyright
WIT Press
Author(s)
N. Nagamine, M. Ukai
Abstract
This research aims to create a simulator that would generate simulated images for use when investigating where to install wayside signals along a railway track. Because the installation of wayside signals requires such considerations on as visibility from the driver’s booth, the current installation measures involve that a staff member holds a simulated emergency wayside signal at a given point and the wayside signal’s visibility is confirmed from a position located in a required distance. These operations require, for both the signal staff and train operation staff, the scheduling of a date for meeting on-site after a lengthy process; this is one reason why it takes considerable time before wayside signals are actually installed. To deal with this problem, we have developed a simulation method that displays a virtual driver’s-eye view of wayside signals on a track by using images of signals overlaid onto the fore view out of the train. This simulation was then used to establish and adjust scheduled wayside signal positions. The simulation uses the fore view of the cabin footage (taken with a camcorder) from the driver’s-eye view and the camera parameters (focal length, height) used therein to predict the three-dimensional space. This three-dimensional space is predicted by using track gauge onscreen as a limiting parameter and solving simultaneous equations. Track gauge is calculated by means of rail extraction techniques used during image processing. The simulation technique cuts back the time needed to study wayside signal positions on-site and requires no restrictions on train schedules or operation. This report describes the algorithms used and explains how to achieve verification by our test data.
Keywords
simulation, installation position, image processing, wayside signal