Resolving Selected Problems in Surface Topography Analysis by Application of the Autocorrelation Function

Abstract

In this paper, the validity of the application of an autocorrelation function for resolving some surface topography measurement problems was presented. Various types of surfaces were considered: plateau-honed, honed with burnished dimples, ground, turned, milled, laser-textured, or isotropic. They were measured with stylus and non-contact (optical) methods. Extraction of selected features, such as form and waviness (defined as an L-surface) and high-frequency measurement noise (S-surface) from raw measured data, was supported with an autocorrelation function. It was proposed to select the analysis procedures with an application of the autocorrelation function for both profile (2D) and areal (3D) analysis. Moreover, applications of various types of regular (available in the commercial software) analysis methods, such as least-square-fitted polynomial planes, selected Gaussian (regression and robust) functions, median filter, spline approach, and fast Fourier transform scheme, were proposed for the evaluation of surface topography parameters from ISO 25178 standards.