Research on Multi-feature Fusion 3D CAD Retrieval Technology Based on Model Segmentation

Accurate and efficient retrieval of existing part models is an important approach for the development of new parts. Currently, retrieval is mostly based on text and two-dimensional information, lacking description of non-geometric information, which affects the accuracy and efficiency of reuse. This article proposes a multi feature fusion similarity retrieval method for 3D solid feature segmentation. Firstly, the part is segmented from the solid features based on a process tree. Secondly, the segmented scatter features are described based on complex networks, and MBD is used to extract information from the solid features. Finally, the Ullmann algorithm is used to solve the common subgraph, achieving part retrieval from both structural and functional aspects.