Research On 3D Reconstruction Algorithm Of Digital Dental Model Based On Feature Measurement

With the continuous development of 3D laser scanning technology,reverse molding technology,and computer-aided design and manufacturing technology,digital impression technology is gradually becoming a research hotspot in the field of prosthodontics and orthodontics.The original data of digital dental model is extremely large and has serious data redundancy,which greatly reduces the processing efficiency of data visualization,storage,transmission and management.This paper mainly focuses on the efficient simplification of massive point clouds and the effective reconstruction of non-uniform point clouds,which is of great significance for the application of complete denture repair and orthodontic treatment.The main research contents of this paper are as follows:1.Aiming at the problem that existing algorithms have non-boundary point misextraction,a boundary point extraction algorithm based on the intersection operation of two boundary point sets is proposed.The boundary point sets of existing algorithms are intersected to effectively constrain non-boundary point misretention.2.To solve the problem that the existing point cloud simplification algorithms are not suitable for the dental model,a point cloud simplification algorithm based on feature significance evaluation is proposed.Through the point cloud feature measurement and analysis of the dental model,the boundary feature points and interdental points are extracted completely,and the feature significance evaluation formula is constructed by selecting the feature measurement factors such as curvature,average projection distance and average spatial distance,so as to realize the accurate extraction of feature points on complex surfaces.An improved spatial subdivision method based on grid method and octree structure is proposed,and the point cloud simplification rule is used to achieve non-uniform simplification of non-feature points.Experimental results show that the proposed algorithm can better maintain the boundary features and complex surface features of the dental model,and effectively avoid the occurrence of holes in the flat region.The proposed algorithm is superior to the existing algorithms in terms of simplification effect and simplification error.3.Aiming at the lack of reference information for radius selection of the ball pivoting algorithm,a point cloud segmentation algorithm based on region growing is proposed.The curvature feature is used to identify and mark the gingival boundary points,and the region growing method is used to realize the effective segmentation of tooth region and gingival region.According to the numerical statistical results of the average point cloud distance of the regional points,the density features of the local point cloud distribution are measured and analyzed to quickly and accurately estimate the number and specific value of the fixed radius.4.Aiming at the problem that the existing ball pivoting algorithm produces holes or redundant triangles when constructing non-uniform point clouds,an improved ball pivoting algorithm based on multi-radius combination is proposed.Based on the relationship between the length of the three triangle sides and its circumradius,the algorithm adaptively estimated the variable radius by using the length relationship between the extended edge and its two vertices participating in all the surrounding constructed mesh edges,and reasonably estimated the fixed radius according to the numerical statistical histogram of the average point cloud distance in the gingival region.Experimental results show that the proposed algorithm can better reconstruct the detailed features of complex surfaces of the dental model,effectively maintain the integrity of the reconstruction results of non-feature regions and reduce the occurrence of holes in the gingival boundary region,and the quality evaluation of the reconstruction results is relatively high.