基于ICT切片图像高抗噪自适应B样条曲线重构  

Highly anti-noise and adaptive b-spline curve reconstruction based on ICT slice image

作　　者：田亚铃[1] 熊鹰[1] 雷华堂[1] 何朝明[1] 李柏林[1] 

机构地区：[1]西南交通大学机械工程学院,成都610031

出　　处：《机械设计与制造》2011年第5期215-217,共3页Machinery Design & Manufacture

基　　金：科技部创新专项科研项目(2009IM032800);四川省科技支撑科研项目(2009GZ0007)

摘　　要：三次B样条曲线重构的关键之一是获得有效的型值点,鉴于ICT切片图像获取的点云数据具有噪声大、数量大、密集等特点,现有基于离散点的曲率的相关算法不利于型值点的选取。利用小波多尺度的技术,采用不同尺度的小波变换估计出的标准误差确定基线段和尺度阀值,提出了一种自动确定型值点和进行曲线重构的基线自适应方法。对重构的样条曲线采用Hausdorff距离计算误差进行评价。通过与现有方法的对比分析,证明该方法能快速重构轮廓曲线,并且能有效抑制噪声,避免了曲线出现畸变和不光滑。Obtaining effective data points is one of key points in cubic B-spline curve reconstruction.Since it is unfavorable for the selection of data points using curvature related algorithms and the point cloud data got from ICT slice images is characterized with large volume of data,high noise and dense,a Baseline Adaptive Algorithm is presented it to get data points for reconstructing curve,baseline and threshold value determined by Wavelet Multi-scale,in which the data points is obtained and curveis reconstructed automatically.Hausdorff distance is adopted to calculate the error of Cubic B-spline Curve Reconstruction.Comparative Analysis with existing methods proves that the method can effectively restrain noise and quickly reconstruct contour curve.

关 键 词：基线自适应 小波多尺度 误差评判 曲线重构 

分 类 号：TH16[机械工程—机械制造及自动化] TP391[自动化与计算机技术—计算机应用技术]