Deviation rectification for dynamic measurement of rail wear based on coordinate sets projection

Chao Wang,Ziji Ma,Yanfu Li,Jiuzhen Zeng,Tan Jin,Hongli Liu
DOI: https://doi.org/10.1088/1361-6501/aa8214
IF: 2.398
2017-09-15
Measurement Science and Technology
Abstract:Dynamic measurement of rail wear using a laser imaging system suffers from random vibrations in the laser-based imaging sensor which cause distorted rail profiles. In this paper, a simple and effective method for rectifying profile deviation is presented to address this issue. There are two main steps: profile recognition and distortion calibration. According to the constant camera and projector parameters, efficient recognition of measured profiles is achieved by analyzing the geometric difference between normal profiles and distorted ones. For a distorted profile, by constructing coordinate sets projecting from it to the standard one on triple projecting primitives, including the rail head inner line, rail waist curve and rail jaw, iterative extrinsic camera parameter self-compensation is implemented. The distortion is calibrated by projecting the distorted profile onto the x–y plane of a measuring coordinate frame, which is parallel to the rail cross section, to eliminate the influence of random vibrations in the laser-based imaging sensor. As well as evaluating the implementation with comprehensive experiments, we also compare our method with other published works. The results exhibit the effectiveness and superiority of our method for the dynamic measurement of rail wear.
engineering, multidisciplinary,instruments & instrumentation
What problem does this paper attempt to address?