Cubic B-Spline Fitting Method Based On Dominant Information

Reconstruction of curves and surfaces with measured data is the core work of reverse engineering.The reconstruction using measured data will make the modeling better.Usually, the curve fitting is the first step and then surface representation can be obtained.The point of this paper is curve fitting and surface reconstruction.By researching the traditional methods,we use Cubic B-spline to fit a set of ordered data. The knot placement is determined by averaging the parameter values of the dominant points, which play an important role in producing better curve approximation.The resulting curve is error-bounded with fewer control points based on energy minimization.In surface construction,one algorithm based on the characteristic is also considered,which constitutes the mapping of characteristics and transfer it to the mapping of parameter.By using our method,we can generate a B-spline curve or surface with less deviation and the dominant information of the data set also appears.Some experimental results demonstrate our approach usefulness and quality.