New ways of using tests on statistical properties of randomly real rough surfaces

Randomly rough real surfaces are characterized by their profile which is described by its statistical moments. A common assumption is that the distribution functions for these profiles are shaped Gaussianly. This article presents some statistical tests to characterize real rough surfaces in a quantitative way, and to assess how and what percent of these probability functions depart from Gaussian functions. Results of several tests are given on various thin films with rough surfaces.