Application of FTP in flaw detection of rail web

Abstract

In this paper, the author aims to use the Fourier transform profilometry, a three-dimensional non-contact optical measurement method, for recovering the rail profile and defects on the rail web. The deformed fringe pattern, caused by the uneven distribution of the surface's height when a Ronchi grating is projected onto the rail's surface, is captured by the CCD camera and handled by Fourier transform, filtering, Fourier inverse transform and phase unwrapping and then the rail profile can be rebuilt. In phase unwrapping, least square algorithm based on the specific characteristic of the rail is used. Simulation is done to manifest the feasibility of the method. Results of the experiment show that in restoring the rail profile and detecting the rail web's defects, Fourier transform profilometry, compared to some traditional surface detection method, is able to achieve a higher accuracy. It shows a great advantage and also provides a better basis and testing demonstration for the further application of the non-contact optical three-dimension measuring method on the rail.