Research On Optimization For Mesh Model Reconstructed By Industrial CT Serial Images And System Development

Industrial CT technology, best non-destructive testing methods, is widely used in the field of reverse engineering, because it can obtain complex internal geometric surface data without damaging the objects. However, due to operator error, equipment aging and other reasons, the industrial CT serial images resolution in z-direction is commonly far lower than the imaging resolution in the xy-plane, which have a bad influence on the subsequent reverse engineering. In order to improve the accuracy of reverse engineering, from the industrial CT serial images to the solid model need to go through the mesh reconstruction, mesh simplification, mesh standardization and so on. This paper focuses on mesh reconstruction, mesh simplification and mesh standardization, the main contents and achievements are listed as follows:① In order to obtain a mesh model, the Marching Cubes algorithm is used to reconstruct the industrial CT serial images. First, determine the relationship and the intersections between the three-dimensional voxels and isosurface, and then construct the isosurface inside all voxels, and finally draw the isosurface to obtain the mesh model. This paper takes the industrial CT serial images of the teapot and the engine block as the research object, the experiment result shows that the industrial CT serial images can be better reconstructed by Marching Cubes algorithm.② The model reconstructed by industrial CT serial images have huge data and the triangular meshes quality is poor. To solve these problems, a simplification method with feature preserving for mesh model was proposed. In this method, the original model was simplified by triangle collapse, and then after the average dihedral angle error of mesh model reached the allowable error, the model was simplified by edge collapse. For triangle collapse, the folding point of one triangle was determined by its normal vector, the Gaussian curvature of each vertex on it and vertices projection method, the collapsing cost of one triangle was determined by the sum of dimensionless number of its local volume error and its dihedral angle error. For edge collapse, the dihedral angle error was introduced in the edge collapse cost of QEM. The experiment result shows that the presented simplification method could generate the feature preserving, high quality and lower geometric error simplified model compared with most of the existing simplification algorithms.③ In order to further improve the triangular meshes quality of the simplified model, a mesh standardization method, combines local mesh surface fitting and intelligent optimization algorithm, based on particle swarm optimization algorithm was proposed. First, using cubic polynomial surface to fit each vertex of the mesh and its local neighbors, and then taking the fitting surface as the search field of improved particle swarm optimization algorithm, taking the overall quality of local triangular meshes as the objective function, using particle swarm optimization algorithm to make the objective function to maximize, and finally calculating the vectors error of vertex before and after adjustment, standardizing the model according to the error. The experiment result shows that the presented standardization method could generate the feature preserving, high quality and lower geometric error optimized model compared with most of the classical standardization algorithms.The research achievements above have been integrated into a prototype system of reverse engineering,and also applied to the inner structures of the parts in reverse design.