基于法向集的点对一致性投票的点云法向估计算法
Point Cloud Normal Estimation Algorithm Based on Normal Set-Based Point Pair Consistency Voting
DOI: 10.12677/aam.2026.159385, PDF,   
作者: 周佳怡：辽宁师范大学数学学院，辽宁 大连
关键词: 点云；法向估计；法向集；Point Cloud； Normal Estimation； Normal Set
摘要: 法向量是点云数据中至关重要的几何属性，广泛应用于特征提取、三维分割、曲面重建等任务。在法向估计中，如何评价两个邻域点的相似性程度对邻域点的筛选与加权是至关重要的。PCV算法利用两个点法向之间的一致性来刻画邻域点的相似性程度，较好地保持了局部邻域的特征，但其使用单一、粗糙的初始输入法向进行一致性度量，使得在尖锐特征附近对邻域点相似性的判断不够精准。针对这一问题，本文提出一种PCV改进算法，通过为尖锐特征点构建法向集，将PCV中单一初始法向间的一致性度量替换为法向集之间的一致性度量。实验结果表明，本文算法在保持点云特征的同时，在不同噪声水平下均取得了更好的法向估计结果。
Abstract: Normal vector is a critical geometric property of point cloud data, and is widely applied in tasks including feature extraction, 3D segmentation, and surface reconstruction. In normal estimation, evaluating the similarity between two neighboring points is of great significance for the selection and weighting of neighboring points. The PCV algorithm characterizes the similarity of neighboring points via the consistency between the normal vectors of two points, which preserves the features of local neighborhoods well. However, it uses a single, rough initial input normal to measure consistency, leading to inaccurate judgment of neighboring point similarity near sharp features. To address this problem, this paper proposes an improved PCV algorithm that constructs a normal set for sharp feature points, and replaces the consistency measurement between single initial normals in PCV with the consistency measurement between normal sets. Experimental results show that while preserving point cloud features, the proposed algorithm achieves superior normal estimation results under different noise levels.
文章引用：周佳怡. 基于法向集的点对一致性投票的点云法向估计算法[J]. 应用数学进展, 2026, 15(9): 199-211. https://doi.org/10.12677/aam.2026.159385

参考文献

[1] Amenta, N. and Bern, M. (1999) Surface Reconstruction by Voronoi Filtering. Discrete & Computational Geometry, 22, 481-504.
https://doi.org/10.1007/pl00009475
[2] OuYang, D. and Feng, H. (2005) On the Normal Vector Estimation for Point Cloud Data from Smooth Surfaces. Computer-Aided Design, 37, 1071-1079.
https://doi.org/10.1016/j.cad.2004.11.005
[3] Dey, T.K. and Goswami, S. (2006) Provable Surface Reconstruction from Noisy Samples. Computational Geometry, 35, 124-141.
https://doi.org/10.1016/j.comgeo.2005.10.006
[4] Alexa, M., Behr, J., Cohen-Or, D., Fleishman, S., Levin, D. and Silva, C.T. (2001) Point Set Surfaces. 2001 Proceedings Visualization, VIS’01, San Diego, 21-26 October 2001, 21-29.
https://doi.org/10.1109/visual.2001.964489
[5] Jones, T.R., Durand, F. and Zwicker, M. (2004) Normal Improvement for Point Rendering. IEEE Computer Graphics and Applications, 24, 53-56.
https://doi.org/10.1109/mcg.2004.14
[6] Ztireli, A.C., Guennebaud, G. and Gross, M. (2009) Feature Preserving Point Set Surfaces Based on Non‐Linear Kernel Regression. Computer Graphics Forum, 28, 493-501.
https://doi.org/10.1111/j.1467-8659.2009.01388.x
[7] Guerrero, P., Kleiman, Y., Ovsjanikov, M. and Mitra, N.J. (2018) PCPNET Learning Local Shape Properties from Raw Point Clouds. Computer Graphics Forum, 37, 75-85.
https://doi.org/10.1111/cgf.13343
[8] Wang, Y., Sun, Y., Liu, Z., Sarma, S.E., Bronstein, M.M. and Solomon, J.M. (2019) Dynamic Graph CNN for Learning on Point Clouds. ACM Transactions on Graphics, 38, 1-12.
https://doi.org/10.1145/3326362
[9] 张杰, 王佳旭, 史路冰, 等. 基于邻域参与的形状感知卷积网络的点云分析[J]. 辽宁师范大学学报(自然科学版), 2022, 45(4): 448-456.
[10] Li, B., Schnabel, R., Klein, R., Cheng, Z., Dang, G. and Jin, S. (2010) Robust Normal Estimation for Point Clouds with Sharp Features. Computers & Graphics, 34, 94-106.
https://doi.org/10.1016/j.cag.2010.01.004
[11] Boulch, A. and Marlet, R. (2012) Fast and Robust Normal Estimation for Point Clouds with Sharp Features. Computer Graphics Forum, 31, 1765-1774.
https://doi.org/10.1111/j.1467-8659.2012.03181.x
[12] Hoppe, H., DeRose, T., Duchamp, T., McDonald, J. and Stuetzle, W. (1992) Surface Reconstruction from Unorganized Points. ACM SIGGRAPH Computer Graphics, 26, 71-78.
https://doi.org/10.1145/142920.134011
[13] Pauly, M., Gross, M. and Kobbelt, L.P. (2002) Efficient Simplification of Point-Sampled Surfaces. 2002 Proceedings of the IEEE Visualization, Boston, 27 October-1 November 2002, 163-170.
https://doi.org/10.1109/visual.2002.1183771
[14] Guennebaud, G. and Gross, M. (2007) Algebraic Point Set Surfaces. ACM Transactions on Graphics, 26, 23.
https://doi.org/10.1145/1276377.1276406
[15] Cazals, F. and Pouget, M. (2003) Estimating Differential Quantities Using Polynomial Fitting of Osculating Jets. 2003 Proceedings of the Eurographics/ACM SIGGRAPH Symposium on Geometry Processing, Aachen, 23-25 June 2003, 177-187.
[16] Mederos, B., Velho, L. and de Figueiredo, L.H. (2003) Robust Smoothing of Noisy Point Clouds. 2003 Proceedings of the SIAM Conference on Geometric Design and Computing, Seattle, 9-13 November 2003, 1-10.
[17] Wang, Y., Feng, H., Delorme, F. and Engin, S. (2013) An Adaptive Normal Estimation Method for Scanned Point Clouds with Sharp Features. Computer-Aided Design, 45, 1333-1348.
https://doi.org/10.1016/j.cad.2013.06.003
[18] Zhang, J., Cao, J., Liu, X., Wang, J., Liu, J. and Shi, X. (2013) Point Cloud Normal Estimation via Low-Rank Subspace Clustering. Computers & Graphics, 37, 697-706.
https://doi.org/10.1016/j.cag.2013.05.008
[19] Liu, X., Zhang, J., Cao, J., Li, B. and Liu, L. (2015) Quality Point Cloud Normal Estimation by Guided Least Squares Representation. Computers & Graphics, 51, 106-116.
https://doi.org/10.1016/j.cag.2015.05.024
[20] Zhang, J., Cao, J., Liu, X., Chen, H., Li, B. and Liu, L. (2019) Multi-Normal Estimation via Pair Consistency Voting. IEEE Transactions on Visualization and Computer Graphics, 25, 1693-1706.
https://doi.org/10.1109/tvcg.2018.2827998
[21] 周佳怡, 张杰. 基于法向集差异的点云法向估计算法[J]. 应用数学进展, 2026, 15(4): 1-13.

为你推荐