Research On 3D Surface Reconstruction And Smoothing

With the development of modern computer technology and testing technology, people put forward more requirements on rapid product design and product quality, for the reconstruction of product CAD model and the model of fairing surface has become a research hotspot of reverse engineering technology. This paper takes the scattered point cloud obtained by non-contact measurement as the research object, in-depth study of 3D surface reconstruction technology, combined with the characteristics of Bezier surface, NURBS surface,using a Bezier surface is constructed by method of transition of NURBS surface, to solve the surface construction of light and accuracy issues in the process of curved surface.In this paper, the main research content includes:point cloud data preprocessing,3D surface reconstruction, surface smoothness, and surface error analysis evaluation.Firstly, describes the method of product surface digitization, analysis of different measurement method to get the characteristics of point cloud data, according to the point cloud data obtained by scanning in this paper put forward a treatment method of point cloud data simplification efficiency. The method is based on loyalty to the original data, combined with the minimum distance and angle deviation method, the conditions that ensure data reduction method, determining the minimum distance, the point cloud data to streamline.Then, based on the data after pretreatment of the cloud point, surface is reconstructed. According to the characteristics of surface reconstruction and appear in the surface reconstruction process fairness and accuracy problem. This paper attempts to use the method of transition curved surface construction of NURBS surface to Bezier surface is to construct surface.Finally, elaborated the fairing of surface evaluation method, and the surface reconstruction and error analysis and evaluation of fairing.