A Study Of Scattered Point Cloud Self-repairing Reconstruction And Slicing Algorithm In 3D Printing

In recent years, surface reconstruction of scattered point cloud has been widely used in the reverse engineering, medicine and health, virtual reality, 3D printing and so on. Due to measurement error existed in the entity model surface and the defects of surface reconstruction algorithm, lead to the model surface is not smooth, and there are voids, multilateral col-linear errors and inconsistent normal vector in the triangular mesh topology, which bring challenges and difficulties to the surface reconstruction and 3D printing, therefore, a smoothing handling and errors repairing is a key to build a complete model. In addition, being a kind of rapid prototyping technology, 3D printing now has a broad application field and huge development space. Layering process is a key step of 3D printing, whose efficiency directly affects the efficiency and quality of 3D printing. Hence the study of the slicing algorithm has become a research hot spot on 3D printing.The original surface reconstruction algorithm based on projection is simple, fast and the reconstruction result is good, but the algorithm has some limitations, which needs to assume that the surface is smooth and the point cloud is uniform, otherwise the surface will be not smooth or has voids, so a point cloud smoothing algorithm based on moving least square method is used to preprocess the point cloud data. In order to repair multilateral col-linear and normal vector inconsistency error, according to the locality principle of projector reconstruction, the normal of the triangles is adjusted during reconstruction and an incident half-edge list is introduced which is effectively used to reduce the appearance of multilateral col-linear. Experimental results show that the algorithm can repair multilateral col-linear error, improve the quality of the model and provides a complete 3D model.On account of the defects of the STL file, which limits the rapid development of 3D printing, through the analysis of the defects of traditional layered algorithms based on STL model and comparing the advantages and disadvantages of STL and AMF model, according to the locality principle of surface reconstruction algorithm and the characteristic of the AMF model, in this thesis a topology slicing algorithm based on surface reconstruction of scattered point cloud is put forward. The topological structure based on the half-edge match is built, the topological information is stored into the AMF file andfinally the layering process based on the topological information is accomplished. Experimental results show that the algorithm can improve the efficiency of the layering. In this thesis, surface reconstruction of scattered point cloud, repairing of model and the layering process of 3D printing are integrated, which could reduce the errors brought by the step-by-step operations.