中文摘要：

受层间距的影响,工业CT数据重建网格模型时会产生阶梯状伪影.为了在生成网格模型的同时去除阶梯状伪影,提出一种新的去除表面伪影的方法.首先建立工业CT点云数据的k-邻域,通过高斯加权的协方差分析估算出点云法矢量,然后使用双边滤波去除点云中的噪声,最后通过二次误差函数拟合点云,使用自适应圆球覆盖方法对点云进行网格化处理.实验结果表明,所提方法能够有效去除阶梯状伪影,并且在合理的精度范围内可以生成高质量的三角网格模型.

英文摘要：

Due to slices distance factor, surface meshes extracted from industrial CT data often contain stair-stepped artifacts. To remove stair-stepped artifacts during mesh generation, a novel removal surface artifacts method was proposed. The k-nearest neighbors of industrial CT point clouds were constructed, the point cloud vector were esti- mated based on Gaussian weight covariance analysis and the noise from point clouds was removed by bilateral filte- ring. The point clouds were fitted by quadric error function and the triangular meshes was managed by adaptive spherical cover method. The experimental results showed that the proposed approach could efficiently remove stair- stepped artifacts and produce high-quality triangular meshes with a reasonable accuracy.