肝脏三维重建中模型平滑和网格细化方法

Research into Model Smoothing and Mesh Refinement in Three-Dimensional Reconstruction of Liver

针对力触觉再现系统对三维模型真实性和准确性的要求,在基于二维医学图像重构肝脏三维模型的过程中,提出了一种基于B样条拟合曲线的模型平滑方法和基于最邻近点搜索插值的模型网格细化方法。前者是以最初获得的轮廓点为控制顶点,拟合出平滑的B样条曲线并重新采样,以此实现对模型的平滑操作。后者是在每一次连线构造三角形面片过程中,取两点之间的几何中心作为插值点,以此实现三维网格模型的细化。实验表明,这两种算法能够有效提高肝脏三维模型的重建效果,增强力触觉再现系统的真实性。

To fulfill the reality and accuracy requirements of 3D model for haptic rendering,which is reconstructed from medical images,this paper proposes a B-spline curve based smoothing method and a mesh refinement method.The former uses the original contour points as control vertices to generate the B-spline curve for model smoothing.The latter utilizes the nearest point searching and interpolating method to refine the triangle meshes for 3D liver reconstruction.Experiments validate and prove that the presented methods can improve the quality of the 3D liver model reconstruction and enhance the performance of haptic rendering system.