Abstract: It has been assumed that roughness, waviness and form constitute the geometrical surface structure. To separate the above features, analog and digital filters are used in surface metrology. At present, the most popular filter is the filter with a transmission characteristic described by the Gaussian curve. The most natural way of implementation of Gaussian filter is to convolve the profile with the filter impulse response. This kind of filter is called finite impulse response filter (FIR filter). An important drawback of this approach is the difficulty in the determination of the mean line for the first and last profile sections, which is referred in literature as the edge effect. However, the registered profile is often too short to reject its fragments. In the paper we propose a method that enables determining the mean line for the whole registered profile. The method involves proper extension of a registered profile on both its ends by fragments being some polynomial functions of length. The coefficients of the polynomials are selected in such a way that the mean square difference between the profile and the determined mean line reaches the minimum. The results of simulation show that the determined mean line has the expected shape if the magnitude of the form profile component is considerably greater than the magnitude of roughness one.
B I B L I O G R A F I A[1] ASME B46.1. Surface texture (surface roughness, waviness, and lay), 1995.
[2] ISO 11562. Geometrical product specification (GPS) – surface texture: profile method – metrological characteristics of phase corrected filters. Geneva: International Organization for Standardization, 1996.
[3] M. Krystek, P.J. Scott and V. Srinvasan Discrete linear filters for metrology, Proc. of the 16th IMEKO Word Congress, Vienna, Austria, Sept. 25-28, 2000.
[4] V. Srnivasan, Discrete morphological filters for metrology, Proc. Of the 6th IMEKO ISMQC Symposium on Metrology for Quality Control in Production, TU Wien, Austria, 1988.
[5] J. Raja, B. Muralikrishnan, Shengyu Fu, "Recent advances in separation of roughness, waviness and form", Precision Engineering, vol. 26, 222-235, 2002.
[6] P.A. Lynn, W. Fuerst, Digital Signal Processing with Computer Application, 2nd ed., John Wiley & Sons, New York, 1994.
[7] S. Brinkmann, H. Bodschwinna, H.W. Lemke, Accessing roughness in three-dimensions using Gaussian regression filtering, Proc. of the 8th Int. Conf. on Metrology and Properties of Engineering Surfaces, Univ. of Huddersfield, UK, 2000. 
Web of Science