改进权重的迭代最近点算法在点云配准中的应用
|
摘要
针对传统迭代最近点算法不具备抗差性的难题,利用迭代最近点算法配准残差的分布规律,综合M估计及选权迭代思想,提出改进权重的迭代最近点配准算法。根据每个点对配准计算出对应的初始权重,然后在附加点对权重的基础上使用选权迭代法计算出满足条件的权重,以达到抵御粗差的目的。结果表明,选权迭代过程能合理改善三维空间转换参数计算的结果,提出的改进算法较适合含粗差点的点云数据的配准。
In this paper we aim to solve the problem that the traditional iterative closet point algorithm is not robust. Using the iterative closet point registration residuals law, the M-estimators and selecting weight iteration, an improved iterative closet point registration algorithm based on the weight of the point cloud is provided. In order to achieve protection against gross errors, using the residuals for each point on the registration calculation to calculate the corresponding initial weight, we use iteration method with variable weights to calculate suitable weight on the basis of additional points of weights. The experimental results indicate that proposed iteration method is capable of improving the effect of registration, and the improved algorithm is suitable for the registration of point cloud with gross errors.
In this paper we aim to solve the problem that the traditional iterative closet point algorithm is not robust. Using the iterative closet point registration residuals law, the M-estimators and selecting weight iteration, an improved iterative closet point registration algorithm based on the weight of the point cloud is provided. In order to achieve protection against gross errors, using the residuals for each point on the registration calculation to calculate the corresponding initial weight, we use iteration method with variable weights to calculate suitable weight on the basis of additional points of weights. The experimental results indicate that proposed iteration method is capable of improving the effect of registration, and the improved algorithm is suitable for the registration of point cloud with gross errors.
引文
[1] 郑莉. 基于结构光的不规则钣金件三维量测[D]. 武汉:武汉大学,2007(Zheng Li. 3D Measurement of Irregular Sheetmetal Parts Based on Structured Light[D]. Wuhan:Wuhan University,2007)
[2] 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
[3] Natasha G, Leslie I,Rzymon R,et al. Geometrically a Table Sampling for the ICP Algorithm[C]. International Conference on 3D Digital Imaging and Modeling, Banff,2003
[4] 郑德华,岳东杰,岳建平.基于几何特征约束的建筑物点云配准方法[J].测绘学报,2008(37):465-468(Zheng Dehua,Yue Dongjie,Yue Jianping. Geometric Feature Constraint Based Algorithm for Building Scanning Point Cloud Registration[J]. Acta Geodaetica et Cartographic Sinica,2008,37(4):465-468)
[5] Gelfand N,Ikemoto L,Rusinkiewicz S,et al. Geometrically Sampling for the ICP Algorithm[EB/OL]. http:www.cs.princeton. edu/gfx/bubs/Genlfand 2003GSS/stabicp.pdf, 2004-05-18
[6] Chen Y,Medioni G. Object Modeling by Registration of Multiple Range Images[J]. Image and Vision Computing,1992,10(3):145-155
[7] Bergevin R,Soucy M,Gagnon H, et al. Toward a General Multi-View Registration Technique[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1996 , 18 (5):540-547
[8] Park S Y,Subbarao M . An Accurate and Fast Point-to-Plane Registration Technique[J].Pattern Recognition Letters, 2003 , 24 (16):2 967-2 976
[9] Mitra N J,Gelfand N,Pottmann H,et al. Registration ofPoint Cloud Data from a Geometric Optimization Perspective [C].The 2004 Eurographics/ACM SIGGRAPH Symposium on Geometry Processing,Nice, 2004
[10] Ko K H,Maekawa T,Patrikalakis N M. An Algorithm for Optimal Free-form Object Matching[J]. Computer-Aided Disgin,2003,35(10):913-923
[2] 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
[3] Natasha G, Leslie I,Rzymon R,et al. Geometrically a Table Sampling for the ICP Algorithm[C]. International Conference on 3D Digital Imaging and Modeling, Banff,2003
[4] 郑德华,岳东杰,岳建平.基于几何特征约束的建筑物点云配准方法[J].测绘学报,2008(37):465-468(Zheng Dehua,Yue Dongjie,Yue Jianping. Geometric Feature Constraint Based Algorithm for Building Scanning Point Cloud Registration[J]. Acta Geodaetica et Cartographic Sinica,2008,37(4):465-468)
[5] Gelfand N,Ikemoto L,Rusinkiewicz S,et al. Geometrically Sampling for the ICP Algorithm[EB/OL]. http:www.cs.princeton. edu/gfx/bubs/Genlfand 2003GSS/stabicp.pdf, 2004-05-18
[6] Chen Y,Medioni G. Object Modeling by Registration of Multiple Range Images[J]. Image and Vision Computing,1992,10(3):145-155
[7] Bergevin R,Soucy M,Gagnon H, et al. Toward a General Multi-View Registration Technique[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1996 , 18 (5):540-547
[8] Park S Y,Subbarao M . An Accurate and Fast Point-to-Plane Registration Technique[J].Pattern Recognition Letters, 2003 , 24 (16):2 967-2 976
[9] Mitra N J,Gelfand N,Pottmann H,et al. Registration ofPoint Cloud Data from a Geometric Optimization Perspective [C].The 2004 Eurographics/ACM SIGGRAPH Symposium on Geometry Processing,Nice, 2004
[10] Ko K H,Maekawa T,Patrikalakis N M. An Algorithm for Optimal Free-form Object Matching[J]. Computer-Aided Disgin,2003,35(10):913-923