Font Size: a A A

Research On Surface Rendering Algorithm Of 3D Reconstruction Of Medical Tomography Image

Posted on:2021-02-08Degree:MasterType:Thesis
Country:ChinaCandidate:Z H DuanFull Text:PDF
GTID:2404330605473013Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
As a technique to provide 3D model of biological tissue,3D reconstruction of medical tomography has been paid much attention.In terms of medical application,3D reconstruction can not only provide 3D model,but also show the spatial position of 3D model.Through human-computer interaction,doctors can easily diagnose diseases and plan surgeries.3D reconstruction algorithm is mainly divided into surface rendering algorithm and volume rendering algorithm.As one of the most widely used surface rendering algorithms,MC algorithm may have holes in the reconstructed 3D model due to its ambiguity,so it is usually necessary to improve the MC algorithm.Based on the principle of MC algorithm,a 3D reconstruction algorithm of surface rendering based on edge growth is proposed in this thesis.The algorithm first selects the seed triangle and then takes the side of the seed triangle as the growth edge to reconstruction.The seed triangles mainly come from the triangles generated by topological configuration 1.According to the position of the growth edge,the gray value of the cube vertex and the threshold value,the thesis determines 17 basic topological configurations in the growing process.In the interpolation of the 3D reconstruction of the entire algorithm,a three-stage interpolation method is proposed,which has good interpolation efficiency and reconstruction accuracy.At the same time,the algorithm has a queue to store the growing edge information,and two arrays to record the processing information and growing information of each cube voxel.In this thesis,C++ language and VTK are used to program the algorithm.The ambiguity of the algorithm is verified by the 3D model generated by the program.The verification results show that the 3D model reconstructed by the algorithm does not have holes generated by ambiguity.In addition,the algorithm proposed in this thesis is compared with the MC algorithm improved by Masala.The results showthat the algorithm proposed in this thesis was always less than Masala's MC algorithm in terms of the number of triangular faces of the reconstruction model;When only the main contour of the model is needed,the quality of the model reconstructed by the algorithm in this paper is better.Finally,this thesis builds a 3D reconstruction system based on VTK and QT.In the application of the system,the seed area can be set through the interface to reconstruct the three-dimensional model connected to the seeds.
Keywords/Search Tags:3D Reconstruction, Surface rendering, Seed, Growth edge
PDF Full Text Request
Related items