3D Reconstruction of Worn Surface Topography Using Shape from Shading

Surface topography analysis plays an important role in tribology. The electron microscope and the optical microscope are typically used to characterize surfaces topography in tribology. Optical microscope is suitable to acquire in situ surface morphology and geometric feature of worn surface. However, two-dimensional (2D) optical image is less intuitive than the three-dimensional (3D) image. In order to visualize the worn surface with 3D image, the shape from shading (SFS) technique has been used to reconstruct the 3D topographic diagram from the single optical image. Tribological experiments on Ti6Al4V worn surfaces have been performed to obtain the 2D optical images, and then 3D reconstruction has been achieved via SFS method. To validate the accuracy of SFS technique, the surface profile transversal, roughness parameters and the areal field parameters of the target area on worn surface from laser scan confocal microscope (LSCM) have been used to compare with the data from SFS. It is found that the largest height deviation of selected feature points in surface profile transversals between SFS technique and LSCM method is no more than 15%. The maximum peak height (R_p) between SFS and LSCM is 0.40 μm with relative deviation around 13.70%. The largest absolute deviation of maximum peak height (S_p) is 0.60 μm with the maximum relative deviation 17.91%. The above result show that the 3D topography of the worn Ti6Al4V surface reconstructed from 2D optical image by SFS technique is comparable to that obtained from LSCM method, which proves the reliability of SFS technique in image processing of worn surfaces.