Form removal aspects on the waviness parameters for steel sheet in automotive applications: Fourier filtering versus polynomial regression

Premium car makers attach great importance to the visual appearance of the painted car skin as an indication of product quality. The "orange peel" phenomenon constitutes a major problem here. It is not only depending on the paint's chemical composition and application method, but also on possible waviness components in the sheet substrate. Therefore one is searching hard for a valuable waviness parameter to quantify the substrate's fitness for purpose. A technically emerging problem is how to remove the form from the measured signal, which is indeed not significant to the orange peel phenomenon. This paper will compare two commonly used approaches: i.e. Fourier filtering versus polynomial regression and will reveal and quantify some common aspects in terms of wavelengths.

http://ift.tt/2sdiWNU