Evaluation of surface roughness based on monochromatic speckle correlation using image processing

The measurement of roughness on machined surfaces is of great importance for manufacturing industries as the roughness of a surface has a considerable influence on its quality and function of products. In this paper, an experimental approach for surface roughness measurement based on the coherent speckle scattering pattern caused by a laser beam on the machined surfaces (grinding and milling) is presented. Speckle is the random pattern of bright and dark regions that is observed when a surface is illuminated with a highly or partially coherent light beam. When the illuminating beam is reflected from a surface, the optical path difference between various wavelets with different wavelength would result in interference showing up as a granular pattern of intensity termed as speckle. The properties of this speckle pattern are used for estimation/quantification of roughness parameters. For measurement of surface roughness, initially the speckle patterns formed are filtered in the spatial frequency domain. The optical technique, namely spectral speckle correlation (autocorrelation) is utilized in this work for the measurement of roughness on machined surfaces. It has been observed that the pattern formed is dependent on the roughness of the illuminated surface. For example, a rough surface (milled) shows a small central bright region with a rapid decrease in intensity towards the edges, while a smooth surface (ground) shows a large central bright region with gradually decreasing intensity towards the edges. The complete methodology and analysis for quantification/estimation of surface finish of milled and ground surfaces based on speckle images that could be implemented in practice, is presented in this paper.