基于法向量和高斯曲率的点云配准算法

石磊; 严利民

Chinese

您当前的位置：

首页 >

文章列表页 >

基于法向量和高斯曲率的点云配准算法

更新时间：2026-09-04

- 基于法向量和高斯曲率的点云配准算法
- Point cloud registration algorithm based on normal vector and gaussian curvature
- 微电子学与计算机 2020年37卷第9期页码：68-72
- 作者机构：
  
  上海大学微电子研究与开发中心,上海,200072
- 作者简介：
- 基金信息：
  
  国家自然科学基金 61674100
- DOI：
  中图分类号：
- 纸质出版：2020
- 稿件说明：
移动端阅览
石磊, 严利民. 基于法向量和高斯曲率的点云配准算法[J]. 微电子学与计算机, 2020, 37(9): 68-72.

SHI Lei, YAN Li-min. Point cloud registration algorithm based on normal vector and gaussian curvature[J]. 2020, 37(9): 68-72.
石磊, 严利民. 基于法向量和高斯曲率的点云配准算法[J]. 微电子学与计算机, 2020, 37(9): 68-72. DOI：

SHI Lei, YAN Li-min. Point cloud registration algorithm based on normal vector and gaussian curvature[J]. 2020, 37(9): 68-72. DOI：

摘要

迭代最近点算法(Iterative closest point

ICP)因配准精度高、适应性强而被广泛使用，但是它容易受高斯噪声和离群点的影响，导致运行速度缓慢、配准精度降低，且需要两片点云具有良好的初始位置，否则会出现局部最优问题.针对以上问题，本文提出了一种新的点云配准方法，利用法向量和高斯曲率进行粗配准，去除无关点同时提供较好的初始位置，再采用基于奇异值分解的ICP算法进行精细配准，并采用斯坦福大学的点云数据集进行了配准实验，结果表明，本文算法能够有效降低高斯噪声和离群点对配准效果的干扰，改善了点云配准的运行效率和配准精度，与传统ICP算法相比，平均配准时间减少了53.5%，配准精度提高了43.2%.

Abstract

Iterative closest point(ICP) algorithm is widely used due to its high registration accuracy and strong adaptability. However

it is easily affected byGaussian noise and outliers

resulting in slow running speedandreduced registration accuracy

whichrequires that the point cloud must havea good initial position

otherwise local optimization problems will occur. In view ofthe above problems

this paper proposes a new point cloud registration method

which uses coarse vector and Gaussian curvature for coarse registration

removes irrelevant points and provides a good initial position

and then uses singular value decomposition based ICP algorithm to refine registration

and use Stanford′s point cloud data to set for registration experiments. The results show that the algorithm in this paper can effectively reduce the interference of Gaussian noise and outliers on the registration effect and improve the operation efficiency and registration of point cloud registration. Compared with the traditional ICP algorithm

the average registration time is reduced by 53.5% and the registration accuracy is increased by 43.2%.

关键词

Keywords

references

VLAMINCK M, LUONG H, PHILIPS W. Multi-resolution ICP for the efficient registration of point clouds based on octrees[C]//2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA). Nagoya, Japan, IEEE, 2017: 334-337.

BESI P J, MCKAY N D. Method for registration of 3-D shapes[C]//Sensor fusion Ⅳ: control paradigms and data structures. International Society for Optics and Photonics. Boston, 1992(1611): 586-606.

LUO W, HAO X, XU C, et al. Coarse-to-fine registration for infrared and visible images of power grid[C]//2017 4th International Conference on Systems and Informatics (ICSAI). Hangzhou, China, IEEE, 2017: 1181-1185.

ALI H, FIGUEROA N. Segmentation and pose estimation of planar metallic objects[C]//2012 Ninth Conference on Computer and Robot Vision. Washington, DC, USA, IEEE, 2012: 376-382.

HE H, WANG H L, SUN L. Research on 3D point-cloud registration technology based on Kinect V2 sensor[C]//2018 Chinese Control And Decision Conference (CCDC). Liaoning, China IEEE, 2018: 1264-1268.

THO T P, THINH N T. Design and development of the sorting system based on robot[C]//2015 15th International Conference on Control, Automation and Systems (ICCAS). Busan, Korea, IEEE, 2015: 1639-1644.

ESMAEILIi A, MARVASTI F. A novel approach to quantized matrix completion using huber loss measure[J]. IEEE Signal Processing Letters, 2019, 26(2): 337-341. DOI: 337

WANG Q, TAN Y, MEI Z. Computational Methods of Acquisition and Processing of 3D Point Cloud Data for Construction Applications[J]. Archives of Computational Methods in Engineering, 20 19: 1-2 1. DOI: 10. 1007/s 1 183 1-0 19-09320-4.

SURESH K V, DINESH R. 3D Point Cloud Registration using A-KAZE Features and Graph Optimization[C]//2019 International Conference on Communication and Signal Processing (ICCSP). Madrid, Spain, IEEE, 2019: 0898-0902.

HE Y P, GU L X. Medical image registration using normal vector and intensity value[C]//2011 International Conference on Intelligent Computation and Bio-Medical Instrumentation. Wuhan, China, IEEE, 2011: 46-49.

LI Y, TONG G, DU X, et al. A single point-based multilevel features fusion and pyramid neighborhood optimization method for ALS point cloud classification[J]. Applied Sciences, 2019, 9(5): 951. DOI: 951

LIU J X, QU A J. The historical process of Gauss's wonderful theorem[J]. Studies in Dialectics of Nature, 2017 (9): 108-113. DOI: 108

LIU H, LIU T, LI Y, et al. Point cloud registration based on MCMC-SA ICP algorithm[J]. IEEE Access, 2019(7): 73637-73648. DOI: 73637

TANG S W, JIA R Y. K-Means cluster algorithm based on improved PSO[J]. Computer Engineering and Applications, 2019 (18): 22. DOI: 22

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于多尺度融合与小目标增强的航拍车辆检测算法

基于MSS-YOLO-MPHands融合架构的装配手部动作识别研究

一种用于极低压boost系统的高效率电荷泵

V型沟槽SiC中子探测器的蒙特卡洛分析

超低功耗实时时钟电路的设计