In this work, we propose a method to estimate basic parameters like the rms roughness and the mean grain size of nanocrystalline thin films on rough substrates. The method is based on the analysis of the power spectral density (PSD) of the surface profile, which allows distinguishing between the two participating components from surface and film. The effectiveness will be demonstrated for thin NiO(x) layers for gas sensing on Al(2)O(3) ceramic substrates, and for protective WC coatings on steel.