Feature-based reverse engineering of mechanical parts

Reverse engineering of mechanical parts requires extraction of information about an instance of a particular part sufficient to replicate the part usi ng appropriate manufacturing techniques. This is important in a wide variet y of situations, since functional CAD models are often unavailable or unusa ble for parts which must be duplicated or modified. Computer vision techniq ues applied to three-dimensional (3-D) data acquired using noncontact, 3-D position digitizers have the potential for significantly aiding the process . Serious challenges must be overcome, however, if sufficient accuracy is t o be obtained and if models produced from sensed data are to be truly usefu l for manufacturing operations. This paper describes a prototype of a rever se engineering system which uses manufacturing features as geometric primit ives, This approach has two advantages over current practice, The resulting models can be directly imported into feature-based CAD systems without los s of the semantics and topological information inherent in feature-based re presentations. In addition, the feature-based approach facilitates methods capable of producing highly accurate models, even when the original 3-D sen sor data has substantial errors.