Fillet Detection on Noisy CAD Meshes Using Curvature Direction Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting fillets in CAD models are inaccurate, especially on noisy meshes, due to reliance on curvature values which are not robust and affected by noise and non-smooth surfaces.
Innovation Solution
A computer-implemented method that determines curves following maximal curvature directions on a mesh, fits these curves with circles, and calculates statistics such as mean fitting error and radius variance to accurately detect fillets, improving robustness and precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If curvature values are used for fillet detection, then the detection process is simple, but the accuracy deteriorates on noisy meshes
Solution Approach 1:
The patent transforms the detection approach from using scalar curvature values to using curvature direction vectors and their statistical distribution. This parameter transformation enables robust fillet detection on noisy meshes by analyzing the orientation patterns of maximal curvature directions rather than relying on noisy curvature magnitudes alone.
Solution Approach 2:
The patent introduces a new dimension of analysis by computing the distribution of maximal curvature directions across the mesh surface. Instead of examining curvature at single points, the method analyzes directional patterns across multiple points and surfaces, adding a spatial distribution dimension that filters out noise and reveals true fillet geometry.
2Productivity
If curvature values are used for fillet detection, then the method is computationally efficient, but reliability deteriorates due to noise sensitivity
Solution Approach 1:
The patent creates a statistical representation (distribution model) of curvature directions that serves as a robust copy of the underlying fillet geometry. This statistical copy is insensitive to local noise in the mesh while preserving the essential geometric characteristics of the fillet, enabling reliable detection even when individual curvature measurements are noisy.
3Measurement precision
If statistical analysis of curvature directions is performed, then detection accuracy improves, but computational complexity increases
Solution Approach 1:
The patent segments the mesh into multiple surfaces and computes curvature direction statistics independently for each surface. This segmentation approach allows parallel computation and reduces the overall computational burden while maintaining high detection accuracy through localized statistical analysis that captures fine geometric details.
Data Source
AI summary
A computer-implemented method including obtaining a mesh representing a segment of an outer surface of a portion of a mechanical part. The method further including determining curves over the mesh that each follows maximal curvature directions of the mesh, fitting each curve with a respective circle, thereby obtaining a set of circles, and calculating a value of one or more statistics of the set of circles. The method then detects whether the mesh is a fillet or not as a function of the value of the one or more statistics.