Computer Vision for Train Tracking System Using Discrete Wavelet Transform

Majid Memarian Sorkhabi^1, , Sadegh Bakhshi¹ and Saeid Bashirian¹

¹Tabriz Urban Railway Organization (TURO), Tabriz, Iran

Cite this paper:
Majid Memarian Sorkhabi, Sadegh Bakhshi and Saeid Bashirian. Computer Vision for Train Tracking System Using Discrete Wavelet Transform. American Journal of Computing Research Repository. 2014; 2(3):53-57. doi: 10.12691/ajcrr-2-3-3

Abstract

Modern Mass Transit Systems must attract travelers in competition with other modes of urban transport such as private cars. In order to be successful, the transit system must be easy to access, give short waiting time for trains, provide a fast and comfortable journey and be safe and reliable. For the system control and operator, knowledge of train position provides the means to supervise and control traffic in the whole line, with a high degree of automation. Using video processing for train position estimation can reduce the cost of system implementation. It can also reduce operating costs by easy support to maintenance activities. In order to improve the vehicle tracking quality in the Tabriz urban railway area, digital video processing and discrete wavelet transform are utilized for detection train position and entry in stations. Firstly all real video frames, which extracted from CCTV system, are decomposed to red, green and blue images. Secondly differences of five consecutive frames are calculated. Then wavelet transform of these images are calculated with symlets wavelet. Finally Wavelet Coefficient Energy of each pixel is compared with pre-selected threshold for estimation of train position. We reach the accuracy of 94.3% in train position estimation when it is entered in stations, and 93.7% accuracy for determination of train entry in real area experience.

Keywords:
video processing train position tracking wavelet transform signaling system

This work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Figures

References:

[1]	Simon, Iwnicki, Handbook of Railway Vehicle Dynamics (Eds), Taylor & Francis, 2006, 11-19.
[2]	C.-Y, Qian, Z.-Z, Han, W.-D, Xie, “Application of FPGA to absolute positioning system of maglev vehicles”. Chinese Journal of Scientific Instrument, vol, 27, no. 6, pp. 884-886, 2006.
[3]	I. Yoda, "Image processing technology for advanced safety to people in railroad transportation - For railroad crossing and station platform," IPSJ Magazine Vol. 48, No. 1, pp. 10-16, Jan. 2007.
[4]	M-S, Kim, “Design of the Active Steering Fuzzy Controller with the Degree of Non-uniformity for a Scaled Railway Vehicle,” The 9th WSEAS International Conference on Robotics, Control and Manufacturing Technology (ROCOM '09), Hangzhou, China, May 20-22, 2009.
[5]	S. Shen, T. X. Mei, Active Steering of Railway Vehicles: A Feed forward Strategy, European Control Conference, 2003.