Sophisticated filtration technique for 3D surface topography data of rectangular area

Surface texture in general consists of microscopic surface roughness, waviness and form geometry. They can be discriminated by surface wavelength limits or cut-off values. Recently, 3D surface texture assessment has become popular even in the industrial world. This means that areal filtration is essentially needed and a robust authorized digital filter for 3D surface topography data is strongly required. The filtering performance may be expressed by the amplitude transmission characteristic and the end effect near the data boundary. A novel areal filtering technique is proposed in our paper, by which we can easily obtain an arbitrary amplitude transmission characteristic with less end effect. This technique is related to frequency-domain analysis instead of a conventional convolution method. Reduction of the end effect can be achieved by extrapolating reasonable data at the points outside a rectangular measuring area. Its validity was established using some examples.