Structure feature recognition technology based on extended attribute adjacency graph

Driven by advances in complex part modeling and intelligent manufacturing, the efficient and accurate recognition of features in 3D models has become a central challenge in computer-aided design (CAD) and computer-aided manufacturing (CAM). This paper proposes a feature recognition method using an extended attribute adjacency graph (EAAG). To overcome the limitations of traditional attribute adjacency graphs, the paper introduces an attribute set that captures extended feature surfaces and their relationships, together with corresponding analytical rules and algorithms. This approach enables the distinct identification and representation of feature surfaces, their geometric and topological properties, and other extended information. Furthermore, a standard feature structure library is established. A feature recognition algorithm leveraging the VF2 subgraph matching method is then introduced. This combination significantly enhances the efficiency and accuracy of feature recognition for complex parts. Experimental results demonstrate that the proposed method is capable of accurately identifying various typical feature structures, thereby providing a reliable foundation for integrated CAD/CAE/CAM design and manufacturing.