<title>In-process roughness characterization of specularly reflecting surfaces using doubly scattered light</title>

Abstract

The statistical properties of speckle patterns that are generated in the Fresnel region, when a rough surface is illuminated with a fully developed static speckle pattern are studied. The intensity autocorrelation function characterizes the roughness of specularly reflecting surfaces. The measuring effect is based on a roughness-dependent spatial intensity modulation of the speckle field which is scattered from a surface under speckle pattern illumination. In addition, anisotropic surfaces give rise to an anisotropy of the speckle modulation phenomenon. The speckle patterns under investigation are first recorded by use of a CCD camera and are then evaluated by digital image processing in order to determine a 2D-autocorrelation function. The main advantage of this approach compared to profilometric and light scattering methods such as angle resolved scattering (ARS) and total integrated scattering (TIS) is its reliability and its capability to in-process applications. The measurement results basically depend on the rms roughness. In comparison with ARS- based measuring principles, the surface autocorrelation length shows only little influence. Furthermore, only a small angular range (less than 5 deg.) of the scattered light distribution needs to be evaluated, so that distances of more than 100 mm between optical setup and rough surface can be realized. Earlier investigations in this field deal with speckle patterns obtained from transmitting isotropic surfaces. In this study reflecting anisotropic surfaces, which are typically produced by mechanical processes such as grinding and turning are taken into consideration. Therefore, a more general theoretical description of the rough surface which covers both, isotropic and anisotropic roughness will be given.© (1999) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.