AN Innovative Solution for Face Identification from Wireframes in Cad Modeling System

Identifying face topology from wireframe profiles is a critical challenge in parametric CAD modeling, especially when dealing with complex wireframe models. Existing techniques often struggle with accuracy and interactive performance, particularly in disorganized wireframes with diverse curve types and complex profiles. This paper presents an Adaptive Generation of Profile (AGP) method that effectively addresses these challenges. Our solution includes three key components: first, we employ adaptive sampling to detect loop planarity and clustering techniques to identify coplanar loops, significantly reducing the search space for containment tests. Second, we introduce a rule-based reasoning approach for a containment test, which swiftly determines nesting relationships among loops. Finally, we implement a Profile Data Cache (PDC) mechanism to cache and map face structures to new entities, minimizing computational overhead. Extensive tests on various wireframe models demonstrate the robustness of our approach, showing significant improvements in both accuracy and processing speed compared to existing methods.