Research On 3D Surface Reconstruction Of Point Cloud Data

With the coming of the information age,reverse engineering technology has been applied more and more widely in virtual reality,archaeology,biomedicine,automobile industry,water conservancy engineering,topographic mapping and other engineering fields.However,with the improvement of the precision of the 3D scanner,the amount of data obtained from the point cloud is very large,and the scattered point cloud data lack the topology structure.Therefore,the two key technologies of reverse engineering are the point cloud data simplification and 3D surface reconstruction,which have been the key issues of people's research.In this paper,three dimensional point cloud data acquired by 3D laser scanner is taken as the research object,mainly focusing on the key technologies of point cloud simplification,triangular mesh generation and mesh smoothing in the process of discrete point cloud reconstruction.The research work and the main achievements at the present stage are summarized as follows:An adaptive point cloud feature point extraction algorithm is proposed to solve the problem that the model feature information is easily lost in the process of point cloud data simplification.By calculating the neighborhood point distance,the approximate curvature value,the data points to the point method and neighborhood point method to set the angle and point cloud parameters,according to the characteristics of the adaptive feature parameters of normal distribution and the mean and variance as the threshold to extract the feature points,do not need to enter the coefficient.Combining the algorithm with spherical parameterization simplification algorithm,it has a good simplified effect for large-scale point cloud data and complex models,and preserves the detail features of the original point cloud model.The experimental results show that the algorithm is simple,stable and time-consuming.Three typical algorithms based on Delaunay triangulation are analyzed in detail.By comparing the advantages and disadvantages,an improved algorithm based on Delaunay growth algorithm is proposed.The adaptive space off-ball strategy is used to find the best advantage of triangle growth,and then the growth and growth of the iterative growth.It is proved by experiment that the triangle mesh reconstructed by this algorithm is fast and the effect is clear.According to the mesh model reconstructed edges exist convex problem,a new model of keeping characteristics of mesh smoothing algorithm is proposed.The algorithm combines the advantages of the Laplasse operator better keep the advantages of triangular regularity and mean curvature method with good accuracy,Laplasse tangential component and the mean curvature method to adjust the grid component the vertex triangulation is more uniform.In this paper,the simulation results of several groups of experiments are compared and analyzed,and the effectiveness and efficiency of the algorithm are proved.