颅骨点云模型的优化配准

由于颅骨的三维点云数据模型复杂且不同人的颅骨差异较小, 对其配准精度要求较高。为了提高颅骨点云模型的配准精度和收敛速度, 提出了一种先粗配准再细配准的配准方法。首先, 对颅骨点云数据模型进行去噪、简化和归一化等预处理; 然后, 通过区域划分、区域配准和求解组合系数以及求解刚体变换等步骤实现区域层次上的颅骨粗配准; 最后, 通过引入动态迭代系数来改进基于旋转角约束的迭代最近点算法, 并采用该改进的ICP算法实现颅骨的细配准, 从而达到精确配准的目的。实验结果表明: 与ICP算法相比, 改进的ICP算法的配准精度和收敛速度分别提高了约30%和50%。证明该种先粗配准再细配准的颅骨点云模型配准方法是一种精度高、速度快的有效颅骨配准算法。

Abstract

As the three dimensional point cloud model for skull is complex and there is little difference in skulls of different people, it has high requirement for the registration accuracy. In order to improve the registration accuracy and the convergence rate of point cloud model for skull, a kind of registration method of coarse regulation first and fine regulation second was proposed. Firstly, the point cloud model for skull should be subject to de-noising, simplification, normalization and other pretreatments; Then, based on regional partition, regional regulation, solving combination coefficient, solving rigid body transformation and other steps, coarse regulation for skull in regional level was realized; Finally, through introducing dynamic iteration coefficient algorithm, the iterative closest point based on the constraint of rotation angle was promoted; and improved ICP algorithm was used to realize the fine regulation for skull in order to achieve the purpose of accurate registration. The experimental result shows that comparing with ICP algorithm, the registration accuracy and the convergence rate of the improved ICP algorithm are separately improved about 30% and 50%. therefore, the kind of registration method of point cloud model for skull of coarse regulation first and fine regulation second is an effective skull regulation algorithm with high accuracy and fast speed.

参考文献

[1] SHUI W Y, ZHOU M Q, DENG Q Q,et al.. Densely calculated facial soft tissue thickness for craniofacial reconstruction in Chinese adults [J]. Forensic Science International, 2016, 266: 573.e1-573.e12.

[2] 张石, 董建威, 余黎煌. 医学图像分割算法的评价方法[J]. 中国图象图形学报, 2009, 14(9): 1872-1880.

ZHANG S, DONG J W, YU L H. The methodology of evaluating segmentation algorithms on medical image [J]. Journal of Image and Graphics, 2009, 14(9): 1872-1880. (in Chinese)

[3] 蔺素珍, 王栋娟, 钟家让, 等. 利用曲率特征虚拟拼接青铜器小碎片的方法[J]. 西安电子科技大学学报(自然科学版), 2016, 43(6): 151-156.

LIN S Z, WANG D J, ZHONG J R,et al.. Approach to reassembling virtual small bronze fragments using the curvature feature [J]. Journal of Xidian University, 2016, 43(6): 151-156. (in Chinese)

[4] 冯筠,陈雨, 仝鑫龙, 等. 三维颅骨特征点的自动标定[J]. 光学精密工程, 2014, 22(5): 1388-1394.

FENG J, CHEN Y, TONG X L,et al.. Automatic feature point extraction for three-dimensional skull [J]. Opt. Precision Eng., 2014, 22(5): 1388-1394. (in Chinese)

[5] LINDEBERG T. Scale Selection Properties of generalized scale-space interest point detectors [J].Journal of Mathematical Imaging and Vision, 2013, 46(2): 177-210.

[6] IZUMIYA S, NABARRO A C, SACRAMENTO A D J. Pseudo-spherical normal Darboux images of curves on a timelike surface in three dimensional Lorentz-minkowski space [J].Journal of Geometry and Physics, 2015, 97: 105-118.

[7] 程国华, 王阿川, 陈舒畅, 等. 多源遥感影像高精度自动配准方法研究[J]. 液晶与显示, 2016, 31(6): 604-612.

CHENG G H, WANG A C, CHEN SH C,et al.. High accuracy-automatic registration method research on multi-source remote sensing image [J]. Chinese Journal of Liquid Crystals and Displays, 2016, 31(6): 604-612. (in chinese)

[8] CAO Z C, MA F L, FU Y L,et al.. A scale invariant interest point detector in Gabor based energy space [J]. Acta Automatica Sinica, 2014, 40(10): 2356-2363.

[9] BESL P J, MCKAY N D. A method for registration of 3-D shapes [J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(2): 239-256.

[10] 王欣, 张明明, 于晓, 等. 应用改进迭代最近点方法的点云数据配准[J]. 光学精密工程, 2012, 20(9): 2068-2077.

WANG X, ZHANG M M, YU X,et al.. Point cloud registration based on improved iterative closest point method [J]. Opt. Precision Eng., 2012, 20(9): 2068-2077. (in Chinese)

[11] LI W M, SONG P F. A modified ICP algorithm based on dynamic adjustment factor for registration of point cloud and CAD model [J]. Pattern Recognition Letters, 2015, 65: 88-94.

[12] MAVRIDIS P, ANDREADIS A, PAPAIOANNOU G. Efficient sparse ICP [J].Computer Aided Geometric Design, 2015, 35-36: 16-26.

[13] DU S Y, LIU J, ZHANG C J, et al.. Probability iterative closest point algorithm for m-D point set registration with noise [J]. Neurocomputing, 2015, 157: 187-198.

[14] DU S Y, ZHENG N N, XIONG L, et al.. Scaling iterative closest point algorithm for registration of m-D point sets [J]. Journal of Visual Communication and Image Representation, 2010, 21(5-6): 442-452.

[15] 吴禄慎, 史皓良, 陈华伟. 基于特征信息分类的三维点数据去噪[J]. 光学精密工程, 2016, 24(6): 1465-1473.

WU L SH, SHI H L, CHEN H W. Denoising of three-dimensional point data based on classification of feature information [J]. Opt. Precision Eng., 2016, 24(6): 1465-1473. (in Chinese)

[16] 韩宝昌, 曹俊杰, 苏志勋. 一种区域层次上的自动点云配准算法[J]. 计算机辅助设计与图形学学报, 2015, 27(2): 313-319.

HAN B CH, CAO J J, SU ZH X. Automatic point clouds registration based on regions [J].Journal of Computer-Aided Design & Computer Graphics, 2015, 27(2): 313-319. (in Chinese)

[17] Wikipedia. Rotation matrix\[EB/OL\]. 2014. http://en.wikipedia.org/w/index.php title=Rotation_matrix&oldid=624815514.

赵夫群, 周明全. 颅骨点云模型的优化配准[J]. 光学精密工程, 2017, 25(7): 1927. ZHAO Fu-qun, ZHOU Ming-quan. Optimization registration of point cloud model of skull[J]. Optics and Precision Engineering, 2017, 25(7): 1927.

颅骨点云模型的优化配准

关于本站 Cookie 的使用提示

全站搜索

颅骨点云模型的优化配准

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索