Parameterization And Fitting Of Unorganized Points In The Reconstruction Of Automobile Body

With the development of product digitization and the fast manufacture technology, the reverse engineering technology has obtained extensive research and application in the shape design of automobile. As the shape design of automobile is much more important key point, so it is the keystone and nodus to shape a high quality smooth surface in the development of a new car. The parameterization of point clouds means that give each point a perfect parameterization in order to make the reconstructed surface smoother. In the sense of that, finding an efficient parameterization for unorganized points has an important impact on improving the fairness of reconstructed surface and the auto body engineering. Our mainly work is as follow:Some typical parameterization methods at home and abroad have been studied and the important meaning of points cloud's parameterization during the surface reconstruction is pointed out.An iterative method for the fair fitting method and the parameterization of unorganized points based on Hoschek algorithm is studied.Based on the four curves fitting for the boundary points using Hoschek iterative method, a suitable Coons surface is constructed as the base surface in order to project the unorganized points onto it along with a certain direction, and the parameterization of unorganized points could be gained from the projected ones.VC++ 6.0 is applied in the mentioned methods for the curve and surface fitting in automobile body, the fitting accuracy and the running time are improved.