基于点云几何特征的错误匹配点对剔除算法

doi:10.19678/j.issn.1000-3428.0049468

计算机工程 ›› 2019, Vol. 45 ›› Issue (4): 163-168. doi: 10.19678/j.issn.1000-3428.0049468

基于点云几何特征的错误匹配点对剔除算法

张少杰,马银中,赵海峰

安徽大学计算机科学与技术学院,合肥 230601

收稿日期:2017-11-28 出版日期:2019-04-15 发布日期:2019-04-15
作者简介:张少杰(1982—),女,讲师、博士研究生,主研方向为模式识别、人机交互、虚拟现实;马银中,硕士研究生;赵海峰,博士研究生。
基金资助:
安徽省教育厅自然科学研究重点项目(KJ2017A016,KJ2016A040)。

Elimination Algorithm of Mismatched Point Pairs Based on Point Cloud Geometric Feature

ZHANG Shaojie,MA Yinzhong,ZHAO Haifeng

School of Computer Science and Technology,Anhui University,Hefei 230601,China

Received:2017-11-28 Online:2019-04-15 Published:2019-04-15

摘要/Abstract

摘要：

针对点云配准过程中存在错误匹配点对的问题,提出一种基于点云几何特征的双阈值剔除算法。依据点云几何特征在刚体变换过程中的平移旋转不变性,在初始匹配点集基础上,借助k近邻方法选取各查询点的近邻点并构成三点对,根据点间距离不变特性完成点对的初步筛选。在此基础上,采用曲面变分描述该三点对所在局部区域的几何特征,通过分析三点对的协方差矩阵,完成匹配点的最终筛选。实验结果表明,该方法可以有效剔除错误匹配点,且具有较高的配准精度。

关键词: 三维点云, 双阈值, 剔除算法, 点间距离, 曲面变分, 近似全等三点对

Abstract:

Aiming at the problem of mismatched point pairs in the point cloud registration process,a double threshold elimination algorithm based on point cloud geometry feature is proposed.According to the geometric characteristics of the point cloud,the translational rotation invariance is obtained during the rigid body transformation process.On the basis of the initial matching point set,the nearest neighbor points of each query point are selected by the k-nearest neighbor method to form a three-point pair.According to the characteristics of the distance between points,the initial pairing is completed.On this basis,the geometrical features of the local region of the three-point pair are described by surface variation,and the final screening of the matching points is completed by analyzing the covariance matrix of the three-point pair.Experimental results show that the method can effectively eliminate the wrong matching points and has higher registration accuracy.

Key words: three-dimensional point cloud, double threshold, elimination algorithm, distance between points, surface variation, approximate congruent three-points pair

中图分类号:

TP399

张少杰,马银中,赵海峰. 基于点云几何特征的错误匹配点对剔除算法[J]. 计算机工程, 2019, 45(4): 163-168.

ZHANG Shaojie,MA Yinzhong,ZHAO Haifeng. Elimination Algorithm of Mismatched Point Pairs Based on Point Cloud Geometric Feature[J]. Computer Engineering, 2019, 45(4): 163-168.

https://www.ecice06.com/CN/Y2019/V45/I4/163

参考文献

［1］TANG P,HUBER D,AKINCI B,et al.Automatic reconstruction of as-built building information models from laser-scanned point clouds:a review of related techniques［J］.Automation in Construction,2010,19(7):829-843.
［2］谷晓英.三维重建中点云数据处理关键技术研究［D］.秦皇岛:燕山大学,2015.
［3］BESL P J,MCKAY N D.A method for registration of 3-D shapes［J］.IEEE Transactions on Patttern Analysis and Machine Intelligence,1992,14(2):239-256.
［4］LEFLOCH D,KLUGE M,SARBOLANDI H,et al.Comprehensive use of curvature for robust and accurate online surface reconstruction［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(12):2349-2365.
［5］SEAL A,BHOWMICK A.Performance analysis of iterative closest point algorithm using modified hausdorff distance［EB/OL］.［2017-10-21］.https://www.irjet.net/archives/V4/i7/IRJET-V4I7565.pdf.
［6］LEE B,LEE D D.Learning anisotropic ICP for robust and efficient 3D registration［C］//Proceedings of International Conference on Robotics and Automation.Washington D.C.,USA:IEEE Press,2016:5040-5045.
［7］WANG D,DU S,ZHOU J,et al.Accurate modified normal based ICP for 3D mapping［C］//Proceedings of International Conference on Internet Multimedia Computing and Service.New York,USA:ACM Press,2016:223-227.
［8］YE Q,YAO Y,GUI P,et al.An improved ICP algorithm for kinect point cloud registration［C］//Proceedings of the 12th International Conference on Natural Computation,Fuzzy Systems and Knowledge Discovery.Washington D.C.,USA:IEEE Press,2016:2109-2114.
［9］YANG J,LI H,CAMPBELL D,et al.Go-ICP:a globally optimal solution to 3D ICP point-set registration［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(11):2241 -2254.
［10］王瑞岩.计算机视觉中相机标定及点云配准技术研究［D］.西安:西安电子科技大学,2015.
［11］ZHOU Q Y,PARK J,KOLTUN V.Fast global registration［C］//Proceedings of European Conference on Computer Vision.Berlin,Germany:Springer,2016:766-782.
［12］史皓良,吴禄慎,余喆琦,等.散乱点云数据特征信息提取算法［J］.计算机工程,2017,43(8):279-283.
［13］秦绪佳,王建奇,郑红波,等.三维不变矩特征估计的点云拼接［J］.机械工程学报,2013,49(1):129-134.
［14］GELFAND N,MITRA N J,GUIBAS L J,et al.Robust global registration［C］//Proceedings of the 3rd Eurographics Symposium on Geometry Processing.Aire-la-Ville,Switzerland:Eurographics Association,2005.
［15］RUSU R B,MARTON Z C,BLODOW N,et al.Persistent point feature histograms for 3D point clouds［EB/OL］.［2017-10-21］.https://ias.in.tum.de/_media/spezial/bib/rusu08ias.pdf.
［16］RUSU R B,BLODOW N,BEETZ M.Fast point feature histograms for 3D registration［C］//Proceedings of IEEE International Conference on Robotics and Automation.Piscataway,USA:IEEE Press,2009:3212-3217.
［17］熊风光,蔡晋茹,况立群,等.三维点云模型中特征点描述子及其匹配算法研究［J］.小型微型计算机系统,2017,38(3):640-644.
［18］MUJA M,LOWE D.Fast approximate nearest neighbors with automatic algorithm configuration［EB/OL］.［2017-10-21］.http://www.cs.ubc.ca/research/flann/uploads/FLANN/flann_visapp09.pdf.
［19］王庆璨.三维点云数据配准算法研究［D］.西安:西安电子科技大学,2015.
［20］NEWCOMBE R A,IZADI S,HILLIGES O,et al.Kinect fusion:real-time dense surface mapping and tracking［C］//Proceedings of the 10th IEEE International Symposium on Mixed and Augmented Reality.Washington D.C.,USA:IEEE Press,2011:127-136.
［21］刘新.三维点云数据的配准算法研究［D］.秦皇岛:燕山大学,2015.
［22］DORAI C,WANG G,JAIN A K,et al.Registration and integration of multiple object views for 3D model construction［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(1):83-89.
［23］GAO Y L,FAN J P.Automatic function selection for large scale salient object detection［C］//Proceedings of the 14th ACM International Conference on Multimedia.New York,USA:ACM Press,2006:97-100.
［24］PAULY M,GROSS M,KOBBELT L P.Efficient simplification of point-sampled surfaces［C］//Proceedings of Conference on Visualization.Washington D.C.,USA:IEEE Computer Society,2002:163-170.
［25］SERAFIN J,GRISETTI G.NICP:dense normal based point cloud registration［C］//Proceedings of International Conference on Intelligent Robots and Systems.Washington D.C.,USA:IEEE Press,2015:742-749.
［26］Fast global registration［EB/OL］.［2017-10-21］.https://github.com/IntelVCL/FastGlobalRegistration/tree/master/dataset.

[1]	顾砾, 季怡, 刘纯平. 基于多模态特征融合的三维点云分类方法[J]. 计算机工程, 2021, 47(2): 279-284.
[2]	霍占强, 王勇杰, 雒芬, 乔应旭. 基于超点图网络的三维点云室内场景分割模型[J]. 计算机工程, 2021, 47(12): 308-315.
[3]	冉超, 方志军, 张彦宇. 基于SDR和改进能量检测算法的频谱感知[J]. 计算机工程, 2020, 46(9): 198-204.
[4]	贺彤,熊风光,韩燮,张元. 一种基于协方差矩阵的点云特征曲线提取算法[J]. 计算机工程, 2018, 44(3): 275-280,286.
[5]	史皓良,吴禄慎,余喆琦,万超. 散乱点云数据特征信息提取算法[J]. 计算机工程, 2017, 43(8): 279-283.
[6]	李论,王一刚,范胜利,白志强,赖建宁. 一种体视显微图像误匹配消除方法[J]. 计算机工程, 2015, 41(5): 285-289.
[7]	李星，郭晓松，郭君斌. 基于多特征融合的前向车辆检测方法[J]. 计算机工程, 2014, 40(2): 203-207.
[8]	谢妍梅，樊臻，张森林. 基于HSI颜色空间的彩色图像边缘检测[J]. 计算机工程, 2013, 39(9): 12-14,19.
[9]	钟顺虹, 何建农. 基于自适应双阈值的SUSAN算法?[J]. 计算机工程, 2012, 38(3): 206-208,211.
[10]	石祥滨, 刘晓娜, 刘芳. 一种双阈值红外行人分割方法[J]. 计算机工程, 2012, 38(12): 5-8.
[11]	綦科, 谢冬青, 刘洁. 基于八叉树空间分割的三维点云模型密写[J]. 计算机工程, 2011, 37(4): 7-9.
[12]	郭永芳, 于明, 黄凯. 基于细菌趋药性的Ostu双阈值图像分割算法[J]. 计算机工程, 2011, 37(22): 8-11.
[13]	王茹, 周明全, 邢毓华. 基于聚类平面特征的三维点云数据精简算法[J]. 计算机工程, 2011, 37(10): 249-251.
[14]	严云洋;郭志波;杨静宇. 基于双阈值的增强型AdaBoost快速算法[J]. 计算机工程, 2007, 33(21): 172-174.

选择文件类型/文献管理软件名称

选择包含的内容

基于点云几何特征的错误匹配点对剔除算法

Elimination Algorithm of Mismatched Point Pairs Based on Point Cloud Geometric Feature

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 14

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于点云几何特征的错误匹配点对剔除算法

Elimination Algorithm of Mismatched Point Pairs Based on Point Cloud Geometric Feature

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 14

编辑推荐

Metrics

本文评价