Abstract

Dynamic tread wear measurement is difficult but significant for railway transportation safety and efficiency. The accuracy of existing methods is inclined to be affected by environmental vibrations since they are highly dependent on the accurate calibration of the relative pose between vision sensors. In this paper, we present a method to obtain full wheel profiles based on automatic registration of vision sensor data instead of traditional global calibrations. We adopt two structured light vision sensors to recover the inner and outer profiles of each wheel, and register them by the iterative closest point algorithm. Computer simulations show that the proposed method is insensitive to noises and relative pose vibrations. Static experiments demonstrate that our method has high accuracy and great repeatability. Dynamic experiments show that the measurement accuracy of our method is about 0.18 mm, which is a twofold improvement over traditional methods.

© 2015 Optical Society of America

Dynamic structured-light measurement for wheel diameter based on the cycloid constraint

Zheng Gong, Junhua Sun, and Guangjun Zhang
Appl. Opt. 55(1) 198-207 (2016)

Correction of rolling wheel images captured by a linear array camera

Jiayuan Xu, Ran Sun, Yupeng Tian, Qi Xie, Ying Yang, Hongdan Liu, and Lei Cao
Appl. Opt. 54(33) 9736-9740 (2015)

Real-time and accurate rail wear measurement method and experimental analysis

Zhen Liu, Fengjiao Li, Bangkui Huang, and Guangjun Zhang
J. Opt. Soc. Am. A 31(8) 1721-1729 (2014)

Standard-wheel-based field calibration method for railway wheelset diameter online measuring system

Yuejian Chen, Zongyi Xing, Yifan Li, and Zhi Yang
Appl. Opt. 56(10) 2714-2723 (2017)

Flexible dynamic measurement method of three-dimensional surface profilometry based on multiple vision sensors

Zhen Liu, Xiaojing Li, Fengjiao Li, and Guangjun Zhang
Opt. Express 23(1) 384-400 (2015)