The application of WTP in 3-D reconstruction of train wheel surface and tread defect

Abstract

In this paper, wavelet transform profilometry (WTP) method is studied to achieve a train wheel surface 3-D topography reconstruction and defects inspection. A structured light illumination is used to generate periodic fringes, and a CCD camera is used to capture the image of deformed fringes caused by a 3-D topography. The Complex Morlet wavelet is used to transform the deformed fringes image. By extracting the maximum value of wavelet transform ridge, distribution of wrapped phase caused by cliffy surface variation is obtained. A reliability algorithm is used for phase unwrapping. Simulation is done to manifest the feasibility of the method, and the maximal deviation of a 3-D reconstructed surface is 0.15mm, so comparing with traditional 3-D reconstruction method, this method achieves a higher accuracy. Validation test on a train wheel shows that this method performs a precise 3-D reconstruction of large dimension object such as wheel surface, while small dimension object such as surface defects is also fine recognized simultaneously. Comparing with light section method, this method achieves a faster 3-D reconstruction.