A Hybrid Interactive Feature Recognition Method Based on Lightweight Model

Lightweight model describes the geometry of an object in terms of triangular meshes, lines and points, which are low level and not suitable for feature recognition. For that reason this paper constructs a feature-oriented lightweight model and represents the lightweight model with Boundary representation (B-rep), which describes the geometry with faces, loops, edges, and vertexes. And then extract features from B-rep using a hybrid interactive feature recognition method, first, pick a face which compose of the feature, and then establish feature´s concave graph by the algorithm of judging adjacency relationship between faces, and then match the concave graph with pre-defined feature template to realize feature recognition, finally, add non-geometric technological information, which is necessary for process planning. A sample application was developed to illustrate the feasibility and validity of the method proposed.