Differential Quantities Calculation And Application For Spatial Curves And Surfaces Of Point Cloud

Acquiring large amounts of point data from real objects has become more conv nient because of modern range sensing technologies and digital scanning devices.the given data points often implicitly determine a smooth underlying surface. The dif-ferential information (such as tangent, normal, curvature and torsion et.), which is the basis of geometric modeling and complex application.and also normal orienta-tion plays a very important role in geometry Processing.In this paper we not only build a framework for differential quantity calculation on spatial curves of point cloud propose a new matching method for given point-set curves,but also improve a new MPA normal estimation methods and propose a normal propagation strategy with a improved adaptive MPA normal estimation method.In chapter 1, we begin with the history and current situation of differential quantity calculation, matching method for given point-set curves,normal propaga tion strategy,and this thesis work. In chapter 2 build a framework for differential quantity calculation on spatial curves of point cloud and propose a new matching method for given point-set curves. We use the differential quantities (including Frenet frame, curvature and torsion) that are computed directly from point cloud to set up a rough and global alignment. With the help of kinematics, we developed a model that minimizes a metric of the tangent distance error to obtain the optimal matching. We also present experiments to indicate that our algorithm is insensi-tive to noise and reaches accurate results. In chapter 4,we Respectively improve the normal orientation MPA method in two and three dimensional cases.Firstly, the weighted PCA method is to proposed. Then,filter the local neighbor points,and (?) timate the normal by the weighted PCA method again. Finally, MPA optimization model and a iterate normal propagation strategy are established to estimate the parameters.which method can estimate the normal accurately,and reduce the com-putation greatly. In chapter 5,on the basis of improved MPA method,we propose a improved normal propagation strategy by adaptive MPA.We end this thesis with a short chapter of conclusions and future works...