Research On Application Of Radial Basis Function In Reverse Engineering

Reconstructing surfaces from digitized points is the pivotal problem of ReverseEngineering. To solve this problem, we can adopt the ideology of VolumeVisualization, regard digitized points as sample data in the process of VolumeVisualization. In this way, a model containing interior information of the object canbe reconstructed and expressed in the form of implicit surface. As the method ofconstruction to be concerned, interpolating digitized points by means of Radial BasisFunction has presented excellent accuracy and smoothness, and been under theresearch by many scholars at home and abroad in recently years. The application ofRadial Basis Function in Reverse Engineering was studied in this thesis. The maincontents are as follows:Θ Surface reconstruction from digitized points based on Radial Basis Functio n (RBF) and Compactly Supported Radial Basis Function (CSRBF) was re searched. The triangle mesh model of surface was extracted by constructin g an implicit surface function. As shown in the experimental result, RBF- based method has better description of details than that of distance-field-b ased method.Θ As the precondition of RBF-based surface reconstruction, normal computation and normal adjustment were investigated. The normal of digitized points was computed use 'micro-tangent plane'. Based on existing algorithms, two new normal adjustment algorithms were put forward and implemented, which greatly improve the speed of computation.Θ Repairing holes in mesh surfaces was discussed as well. A novel hole- repairing algorithm, which can guarantee the consistency between the holes and the original surface, was put forward. The algorithm can be used to repair different kinds of holes in mesh surfaces.Θ Through extracting iso-surface of volumetric data field, the offset surface of arbitrary topology mesh was obtained. The most prominent merit of this method is that the problem of self-intersection can be avoided, which has important practical value in rapid prototyping manufacturing and numerical controlledmachining.