The nitration process influences the mechanical and chemical properties of steel and changes the near-surface characteristics. The nitrided surfaces are less sensitive to corrosive fluids and show a better stability against abrasion. Unfortunately, during treatment pores emerge at the surface. In general this is not desired, since the pores reduce the wear stability. The change of the near-surface characteristics also leads to a remodeling of the surface topography. For example, ground, smooth surfaces show an increased but isotropic roughness after nitration. During the recent years, various speckle techniques for an in-process characterization of surface topography have been improved significantly. One of these promising techniques is the method of trichromatic speckle autocorrelation. Its measuring principle is based on trichromatic light scattering and enables to determine an integral parameter of the surface roughness by the evaluation of the speckle elongation. Especially in the case of nitration, where the specimen is located in a stove filled with ammonia at a temperature of 580 degrees Celsius, this technique offers an in-processing monitoring of surface topography changes from outside the stove. In this paper, the in-process characterization of surface topography by speckle autocorrelation will be introduced. In this context an algorithm has been developed, which allows to estimate the position of the optical axis within the speckle pattern and therefore to determine the surface roughness as well as the local inclination of isotropic surfaces. An important goal of the current research is to realize a reliable process control based on the speckle autocorrelation, that is necessary to produce nitrided surfaces without pores in the compound layer and with good abrasive and corrosive resistance.