Surface Topography Characterization via Two-Step Polynomial Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current surface topography characterization techniques face limitations in accurately measuring large surfaces, as they often misinterpret undesirable topographic contributions for pattern variations, leading to incomplete data representation and loss of critical surface features.
Innovation Solution
A method involving a two-step adjustment process, including data filtering and coefficient determination, to separate and remove undesirable topographic contributions from actual surface features, allowing for higher-order adjustments without affecting pattern data, using equations of varying orders and optional resolution reduction for computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If topography measurements are carried out on large surfaces (greater than 1mm²), then the representativeness of the topographic information is improved, but undesirable topographic contributions (such as support non-flatness) are accentuated
Solution Approach 1:
The patent segments the topographic data processing into multiple steps: first adjusting the entire measurement to remove gross undesirable contributions, then dividing the surface into zones, adjusting each zone separately, and combining results. This segmentation allows removal of support non-flatness effects while preserving local pattern information.
Solution Approach 2:
The patent applies different adjustment treatments to different zones of the measured surface. Each zone is identified and adjusted according to its specific characteristics, allowing local preservation of pattern information while removing undesirable contributions. The adjustment parameters vary locally across the surface.
2Measurement precision
If polynomial adjustment is applied to remove undesirable topographic contributions, then the accuracy of topography characterization is improved, but pattern variations are also removed
Solution Approach 1:
The patent divides the surface into multiple zones before applying polynomial adjustment. By segmenting the measurement area, the adjustment is applied locally rather than globally, which prevents removal of pattern variations that occur in different regions. Each zone's patterns are preserved while undesirable contributions are removed.
Solution Approach 2:
Different polynomial adjustments are applied to different zones of the surface. The adjustment parameters are determined locally for each zone, allowing preservation of local pattern characteristics while removing undesirable topographic contributions. This local quality approach prevents uniform removal of all variations including patterns.
3Area of stationary object
If measurement acquisition size is increased to capture more surface area, then the overall topography representation is improved, but the vertical and lateral resolution requirements become more demanding
Solution Approach 1:
The patent segments the large measurement area into smaller zones that can be processed with appropriate resolution. This segmentation allows the use of standard resolution measurement equipment while still characterizing large surfaces, as each zone is measured and processed independently with sufficient detail.
Solution Approach 2:
The patent processes topographic data by dividing the three-dimensional surface information into multiple two-dimensional zones. This dimensional decomposition allows efficient processing of large areas while maintaining resolution requirements for each zone, transforming a difficult 3D large-area problem into manageable 2D zone problems.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
Method for characterizing the topography of a surface (203), comprising the implementation of the steps of: - measurement (102) of the topography of the surface; - first fitting (104) of the data from the measurement of the topography of the surface, carried out from a first equation of order m; - height filtering (106) of the data from the implementation of the first fitting; - determination (110) of the fitting coefficients of the data from the implementation of the filtering, carried out from a second equation of order n; - second fitting (112) of the data from the implementation of the first fitting, carried out from the second equation of order n and the coefficients previously determined; n and m being positive integers greater than or equal to 1 and such that n ≥ m.