CAD-based computer vision for 3-D object recognition

As CAD/CAM systems contribute to the integration of manufacturing systems, the idea of utilizing CAD as a geometric modeling tool for model-based object recognition motivates the research of building a CAD-based object recognition system. This research proposes a methodology which consists of three subtasks for solving the problem of CAD-based object recognition: (1) transformation from CAD to graph-based representations, (2) transformation from images to graph-based representations, and (3) the matching of the two representations to identify the object in the image. A similarity measure between attributed-graphs is proposed to perform inexact structural matching between attributed-graphs for approximate reasoning of object faces. The CAD database is organized in a hierarchical fashion which helps the matching and verification of higher level features. This system is interfaced to I-DEAS Solid Modeling.