三维面部数据采集与NURBS曲面重构  

3D face data acquisition and NURBS surface reconstruction

作　　者：孟凡文[1,2] 吴禄慎[1] 罗丽萍[1] 

机构地区：[1]南昌大学机电工程学院,江西南昌330031 [2]济宁职业技术学院机电工程系,山东济宁272037

出　　处：《激光与红外》2010年第3期334-338,共5页Laser & Infrared

基　　金：国家自然科学基金项目(No.59965003);江西省教育厅科学技术研究项目(No.GJJ08037);山东省高等学校科技计划项目(No.J09LD53)资助

摘　　要：为获取面部三维点云,并转换为高精度NURBS曲面,建立了面部图像采集与处理系统,并研究其中的点云自动拼接和曲面重构方法。首先,采用双光栅三维扫描仪将正弦光栅投射到人的面部,将被面部调制的变形光栅转换成两片分立点云。然后,通过标定块对测量系统进行标定,利用协方差求取旋转矩阵和平移向量实现点云自动拼接获取完整面部点云,并生成三角面片。最后,通过检测曲率、生成四边形网格、建立UV参数线等处理,进行曲面重构生成高精度NURBS四边域面部曲面,并对重构结果进行误差分析。结果表明:所重构面部曲面符合G1连续,标准偏差为0.009222 mm。本文所采用的高精度的面部曲面重构方法可用于复杂曲面的三维检测与重构中。In order to obtain 3D point clouds of the face, and convert it to precise NURBS surface, the face image acquisition and processing system is established, and its applied algorithms, such as automatic point clouds registration and surface reconstruction are investigated. First, sinusoidal fringe is projected to the face by double fringe 3D scanner, and fringe image is modulated by the face profile. Two separated point clouds of the face are obtained by the conversion of fringe. Then the system is calibrated by the calibration block. Automatic registration of two point clouds of the face is implemented by calculating the rotation matrix and translation vector with coefficient analysis. Triangle patches of the face are obtained. Finally NURBS face surface reconstruction with high precision is implemented by detecting curvature, generating quadrilateral meshes, building UV parameter lines and surface reconstruction, and recon- struction error is analyzed. It shows that the obtained face surface meets the requirement of G1 continuity, and the standard error of the surface reconstruction is only 0. 009222 mm. The reconstruction method used for the face surface can also be utilized in 3D detecting and surface reconstruction of complex surfaces.

关 键 词：逆向工程 面部数据采集 点云拼接 曲面重构 重构误差 

分 类 号：TN911.74[电子电信—通信与信息系统]