Midsurface Extraction of Variable-Thickness Thin-Walled Parts Using an Oriented Signed Distance Field
Advances in Computer, Signals and Systems · 2026 · 10 (1)
DOI: 10.23977/acss.2026.100116 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advances in Computer, Signals and Systems, 2026.
Abstract (Originalsprache)
: Mid-surface extraction of thin-walled parts is a key technique for CAD dimensional reduction modeling and structural analysis. Particularly for variable-thickness structures, traditional offset-based methods struggle to generate stable and topologically correct mid-surface models. To address this issue, this paper proposes an automatic mid-surface extraction method for variable-thickness thin-walled parts based on an oriented signed distance field (SDF). First, the SDF of the model is constructed, and local minimum points are extracted to form a set of mid-surface candidate points. Subsequently, ray-casting– based face pairing relationships are employed to group matched face pairs. On this basis, geometric consistency filtering is performed using distance symmetry and normal opposition constraints to refine the mid-surface points. For constant-thickness and variable-thickness regions, offset generation and boundary-constrained surface fitting strategies are adopted, respectively, to complete mid-surface reconstruction. Experimental results demonstrate that the proposed method achieves good stability and geometric accuracy across models with different thickness characteristics.
Zitieren
APA
(2026). Midsurface Extraction of Variable-Thickness Thin-Walled Parts Using an Oriented Signed Distance Field. Advances in Computer, Signals and Systems, 10(1). https://doi.org/10.23977/acss.2026.100116
Vancouver
Midsurface Extraction of Variable-Thickness Thin-Walled Parts Using an Oriented Signed Distance Field. Advances in Computer, Signals and Systems. 2026;10(1). doi:10.23977/acss.2026.100116
BibTeX
@article{euvimed2026,
title = {Midsurface Extraction of Variable-Thickness Thin-Walled Parts Using an Oriented Signed Distance Field},
journal = {Advances in Computer, Signals and Systems},
year = {2026},
volume = {10},
number = {1},
doi = {10.23977/acss.2026.100116}
}
EUVIMED