Overview
EN ISO 16610-20:2015 - "Geometrical product specifications (GPS) - Filtration - Part 20: Linear profile filters: Basic concepts" defines the fundamental concepts for linear profile filters used in surface texture and profile filtration. Published by CEN/ISO, this part of ISO 16610 establishes the theoretical basis for linear filtering of 1‑D profiles, including definitions, representations, and required filter characteristics used in metrology and quality assurance.
Key topics and requirements
- Definition of linear profile filters: filters that separate a profile into long-wave (form/ waviness) and short-wave (roughness) components and satisfy linearity (F(aX + bY) = aF(X) + bF(Y)).
- Phase-correct (linear phase) filters: filters that avoid phase shifts producing asymmetric distortions of the profile.
- Weighting function and transmission characteristic: the weighting function defines how nearby points contribute to the mean line; its Fourier transform yields the filter’s transmission characteristic (attenuation vs. wavelength).
- Cut-off wavelength: the wavelength at which 50% of a sinusoidal component’s amplitude is transmitted - used as the principal identification and nesting index for linear filters.
- Discrete representation: practical filters operate on sampled profiles; the standard requires discrete representations of data (vectors), weighting functions, and filters (square matrices).
- Filter equations and convolution: linear filtering may be expressed as convolution when the kernel is symmetric and spatially invariant; discrete convolution and matrix formulations are covered.
- Filter banks and multiresolution analysis: structured sets of high‑ and low‑pass filters enable decomposition of profiles into multiple scales (resolutions).
- Conformance requirements: filters conforming to this part must satisfy articulated properties in sections covering equations, discrete convolution, transfer functions and filter bank structures.
Applications and users
This standard is essential for:
- Metrology and quality engineers implementing or validating surface‑texture measurement and analysis methods.
- Manufacturers of profilometers and surface-measurement software that implement filtration algorithms (Gaussian, spline, wavelet variants).
- Research labs and OEMs developing process control for machining, coatings, and tribological surfaces.
- Standards bodies and certification laboratories ensuring traceable, comparable surface‑texture measurements across instruments and industries.
Practical uses include robustly separating roughness, waviness and form; specifying cut-off wavelengths for measurement traces; and implementing compliant digital filters (convolution/matrix) in measurement software.
Related standards
- ISO 16610‑1 (Overview and basic concepts)
- Other ISO 16610 parts: 21 (Gaussian filters), 22 (Spline filters), 28 (End effects), 29 (Spline wavelets) - all expand on filter types and implementation details within the GPS filtration framework.
Keywords: ISO 16610-20, linear profile filters, GPS filtration, weighting function, cut-off wavelength, transmission characteristic, surface metrology, filter bank, discrete convolution.