Study Of Fast Surface Reconstruction And Smoothing For Triangular Meshes

With the development of computer graphics and CAD, geometry modeling is an importantresearch field in the computer graphics. This thesis not only reviews the technology of surfacesreconstruction and mesh smoothing, but also presents two new methods.A fast surface reconstruction algorithm based on Power Crust algorithm is proposed forunorganized points. From sufficient dense sample point set, Power Crust algorithm can reconstructa triangular mesh that is topologically valid and convergent to the original surface. However, thealgorithm is limited in the practical application because of its long running time. A non-uniformdown sampling method with border-estimation is proposed to resample the input data set accordingto local feature before reconstruction. Some instances indicate that this method has greatlypromoted the speed of using scatter data points to finish surface reconstruction, and reallyreconstructed surface model.The paper also proposes a hybrid feature-preserving smoothing method for triangular meshes,which maintians merits of weighted median and Laplacian method. In smoothing process, thefeature vertices weighted differently. For feature vertices the displacement is small and for commonones the displacement is big. After smoothing process, features of the original mesh can beperfectly preserved. Compared with existing methods, this method can efficiently eliminate noisesand preserve features of the original meshes.