An evaluation of surface roughness parameters measurement using vision-based data

This paper presents a methodology for using machine vision data to acquire reliable surface roughness parameter measurement. Stylus-based measurements were acquired and compared to vision-based measurements using standard and non-standard roughness parameters. Two light reflection models namely Intensity-Topography Compatible (ITC) model and Light-Diffuse model were adopted and applied to interpret acquired vision data and to enable suitable computation of roughness parameters. Results showed that the ITC model gives more superior results compared to the Light-Diffuse model with remarkable close values to those acquired by the traditional stylus-based data of all roughness parameters but the Skewness parameter (Rsk). In the case of the Skewness parameter (Rsk), however, it was shown to provide highly different values from those acquired by stylus technique. The overall acquired results indicate that vision systems are a valid source of data to obtain both amplitude and spacing roughness parameters with confidence using the proposed methodology. It is expected that these results would encourage further developments in the area to achieve commercial 3D vision-based roughness measurement systems for industrial use.