
基本信息:
- 专利标题: 基于体素模型的影像重建四边形网格方法及系统
- 申请号:CN201811044724.0 申请日:2018-09-07
- 公开(公告)号:CN109285223B 公开(公告)日:2019-05-31
- 发明人: 郑永川 , 王若梅 , 周凡 , 林谋广 , 林格
- 申请人: 中山大学
- 申请人地址: 广东省广州市海珠区新港西路135号
- 专利权人: 中山大学
- 当前专利权人: 中山大学
- 当前专利权人地址: 广东省广州市海珠区新港西路135号
- 代理机构: 广州三环专利商标代理有限公司
- 代理人: 胡枫
- 主分类号: G06T17/20
- IPC分类号: G06T17/20
The invention discloses an image reconstruction quadrilateral mesh method based on a voxel model. The method includes S1, converting CT images to text formats and performing smoothing processing; S2,selecting that gray value and using the interpolation method to extract the equivalent point as the point cloud data; S3, constructing an initial voxel model according to the stacked cube method, optimizing the voxel model, and extracting an outer surface of the voxel model by depth-first searching; S4, calculating a normal vector of each vertex on the outer surface of the voxel model, and searching a point cloud point closest to the normal vector as a mapping point of the vertex through a domain searching method; S5, adjusting the vertices of the quadrilateral mesh to the center of gravity, remapping the vertices, and performing mesh collapse processing on the quadrilateral mesh. The invention also discloses an image reconstruction quadrilateral mesh system based on a voxel model. By adopting the invention, the quadrilateral mesh can be directly reconstructed from the point cloud, and different voxel cube sizes can be selected to control the fineness of the generated quadrilateral mesh.
公开/授权文献:
- CN109285223A 基于体素模型的影像重建四边形网格方法及系统 公开/授权日:2019-01-29
IPC结构图谱:
G | 物理 |
--G06 | 计算;推算;计数 |
----G06T | 一般的图像数据处理或产生 |
------G06T17/00 | 三维(3D)模型,例如3D目标的数据绘图 |
--------G06T17/20 | .有限文件产生,如线框表面的绘图 |