基于四元数最少特征点的相机位姿估计

doi:10.3969/j.issn.1000-1158.2020.05.07

摘要
图/表
参考文献(23)
相关文章 (15)

全文: PDF (951 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要为解决相机位姿估计的问题,提出了一种基于四元数最少特征点的相机位姿估计算法。在相机拍摄的二维图像中检测并匹配特征点,根据特征点坐标与约束条件建立多项式系统,通过求解对应的矩阵方程来求解多项式系统。用四元数表示相机的旋转,避免了求解中相机旋转与平移相互纠缠的问题,当2个摄像机视图之间的平移为零时能够很好地进行位姿求解。求解中结合5点算法对求解原理进行了详细推导,并进行抗噪声测试。测试中随着匹配的特征点数增加,算法平均估计误差范围不超过2%。利用KITTI数据集测试算法的实用性,并与其他几种方法进行结果比较。结果显示,相同条件下算法将估计精度提高了24.5%以上,体现出良好的估计精度。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	侯培国
	赵梓建
	宋涛

关键词 ：计量学, 相机位姿估计, 四元数, 5点算法, 计算机视觉

Abstract：In order to solve the problem of camera pose estimation, a algorithm based on quaternion from least feature points was proposed. A polynomial system, established by constraints and feature points which were detected and matched in the two-dimensional image, was solved by corresponding matrix equation. Using quaternion to represent the rotation of the camera, which avoided the entanglement between rotation and translation in the process of solving. When the translation between two views is equal to zero, the problem can be solved well. The algorithm was deduced in detail by combining five-point algorithm and the anti-noise test was carried out. In the test, as the number of matched feature points increases, the average estimated error range of the algorithm does not exceed 2%. The KITTI dataset was used to test the usefulness of the algorithm and compare the results with several other methods. The results showed that under the same conditions, the algorithm improves the estimation accuracy by more than 24.5%, which shows a good estimation accuracy.

Key words： metrology camera pose estimation quaternion five point algorithm computer vision

收稿日期: 2019-04-22 发布日期: 2020-05-14

PACS:

TB96

基金资助:国家自然科学基金(61601400)

通讯作者: 赵梓建(1994-),山西原平人,燕山大学硕士研究生,主要研究方向为虚拟现实系统与视觉定位。Email: 287611300@qq.com E-mail: 287611300@qq.com

作者简介: 侯培国(1968-),男,山东郯城人,燕山大学教授,博士,博士生导师,主要研究方向为计算机视觉与智能控制。Email: pghou@ysu.edu.cn

引用本文:

侯培国,赵梓建,宋涛. 基于四元数最少特征点的相机位姿估计[J]. 计量学报, 2020, 41(5): 551-557.
HOU Pei-guo,ZHAO Zi-jian,SONG Tao. Camera Pose Estimation Based on Quaternion from Least Feature Points. Acta Metrologica Sinica, 2020, 41(5): 551-557.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2020.05.07 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2020/V41/I5/551

［1］陈丹. 计算机视觉技术的发展及应用［J］. 电脑知识与技术, 2008, 4(8): 2449-2450+2452.
Chen D. The Development and Application of Computer Vision Technology ［J］. [WTBX][STBX]Computer Knowledge And Technology[STBZ][WTBZ], 2008, 4(8): 2449-2450+2452.
［2］张梅, 文静华. 浅谈计算机视觉与数字摄影测量［J］. 地理空间信息, 2010, 8 (2): 17-20.
Zhang M, Wen J H. Simple Discussion on Computer Vision and Digital Photography ［J］. [WTBX][STBX]Geospatial Information[STBZ][WTBZ], 2010, 8 (2): 17-20.
［3］章炜. 机器视觉技术发展及其工业应用［J］. 红外, 2006, 27(2): 11-17.
Zhang W. Development of Machine Vision and Its Industrial Applications ［J］. [WTBX][STBX]Infrared[STBZ][WTBZ], 2006, 27(2): 11-17.
［4］王俊, 朱利. 基于图像匹配-点云融合的建筑物立面三维重建［J］. 计算机学报, 2012, 35(10): 2072-2079.
Wang J, Zhu L. 3D Building Facade Reconstruction Based on Image Matchine-point Cloud Fusing ［J］. [WTBX][STBX]Chinese Journal of Computers[STBZ][WTBZ], 2012, 35(10): 2072-2079.
［5］钱新恩, 夏朝猛. 计算机视觉在工业机器人上的应用［J］. 计算技术与自动化, 2006, 25(1): 114-116.
Qian X N, Xia C M. Application of Computer Vision on Industrial Robot ［J］. [WTBX][STBX]Computing Technology and Automation[STBZ][WTBZ], 2006, 25(1): 114-116.
［6］程淑红, 高许, 程树春, 等. 基于计算机视觉的运动车辆检测［J］. 计量学报, 2017, 38(3): 288-291.
Cheng S H, Gao X, Cheng S C, et al. Moving Vehicle Detection Based on Computer Vision ［J］. [WTBX][STBX]Acta Metrologica Sinica[STBZ][WTBZ], 2017, 38(3): 288-291.
［7］陈春妃, 陈琼, 汤勇明. 基于双目视觉的增强现实模型系统设计［J］. 电子器件, 2016, 39(2): 248-252.
Chen C F, Chen Q, Tang Y M. Design of Augmented Reality Evaluation SystemBased on Binocular Vision ［J］. [WTBX][STBX]Chinese Journal of Electron Devices[STBZ][WTBZ], 2016, 39(2): 248-252.
［8］张立国, 李晓松, 肖磊, 等. 基于单目视觉的四旋翼飞行器目标跟踪算法研究［J］. 计量学报, 2018, 39(3): 342-347.
Zhang L G, Li X S, Xiao L, et al. Research on Target Tracking Algorithm of the Quadrotor Aircraft Based on Monocular Vision ［J］. [WTBX][STBX]Acta Metrologica Sinica[STBZ][WTBZ], 2018, 39(3): 342-347.
［9］扈立超, 史再峰, 庞科, 等. 用于图像匹配的改进Harris特征点检测算法［J］. 计算机工程, 2015, 41(10): 216-220.
Hu L C, Shi Z F, Pang K, et al. Improved Harris Feature Point Detection Algorithm for Image Matching ［J］. [WTBX][STBX]Computer Engineering[STBZ][WTBZ], 2015, 41(10): 216-220.
［10］Philip J. A Non-Iterative Algorithm for Determining All Essential Matrices Corresponding to Five Point Pairs ［J］. [WTBX][STBX]The Photogrammetric Record[STBZ][WTBZ], 1996, 15(88): 589-599.
［11］陈泽志, 吴成柯, 刘勇. 基础矩阵估计的加权归一化线性算法［J］. 软件学报, 2001, 12(3): 420-426.
Chen Z Z, Wu C K, Liu Y. A Weighted Normalization Algorithm for Estimation of Fundamental Matrix ［J］. [WTBX][STBX]Journal of Software[STBZ][WTBZ], 2001, 12(3): 420-426.
［12］陶兆胜, 屠大维. 基于图像几何静矩的机器视觉基础矩阵估计［J］. 仪器仪表学报, 2011, 32(8): 1776-1780.
Tao Z S, Tu D W. Estimation of fundamental matrix based on geometrical static moment of image for machine vision application ［J］. [WTBX][STBX]Chinese Journal of Scientific Instrument[STBZ][WTBZ], 2011, 32(8): 1776-1780.
［13］David Nistér. An efficient solution to the five-point relative pose problem ［J］. [WTBX][STBX]IEEE transactions on pattern analysis and machine intelligence[STBZ][WTBZ], 2004, 26(6): 756-770.
［14］Li H, Hartley R. Five-Point Motion Estimation Made Easy ［C］// IEEE.18th International Conference on Pattern Recognition. 2006.
［15］Kukelova Z, Bujnak M, Pajdla T. Automatic Generator of Minimal Problem Solvers ［C］// European Conference on Computer Vision. 2008.
［16］Wu F C, Hu Z Y, Duan F Q. 8-point algorithm revisited: factorized 8-point algorithm ［C］// Tenth IEEE International Conference on Computer Vision. 2005.
［17］王彤, 马建仓, 秦涛, 等. 基于旋转四元数的姿态解算算法［J］. 弹箭与制导学报, 2014, 34(3): 15-16+20.
Wang T, Ma J C, Qin T, et al. An Attitude Calculating Algorithm Based on Rotation Quaternion ［J］. [WTBX][STBX]Journal of Projectiles. Rockets, Missiles and Guidance[STBZ][WTBZ], 2014, 34(3): 15-16+20.
［18］Kalantari M, Jung F, Guedon J P, et al. The Five Points Pose Problem: A New and Accurate Solution Adapted to, any Geometric Configuration［M］. Advances in Image and Video Technology. Springer Berlin Heidelberg[STBZ][WTBZ], 2009.
［19］Kneip L, Siegwart R, Pollefeys M. Finding the Exact Rotation between Two Images Independently of the Translation ［C］// European Conference on Computer Vision. Springer, Berlin, Heidelberg, 2012.
［20］钟键. 基于对极几何的图像匹配研究［D］. 长沙:中南大学, 2010.
Zhong J. Research on image matching basedon epipolar geometry［D］. Changsha: Central South University, 2010.
［21］
Stewenius H, Engels C, Nistér D. Recent developments on direct relative orientation［J］. [WTBX][STBX]ISPRS Journal of Photogrammetry and Remote Sensing[STBZ][WTBZ], 2006, 60 (4): 284-294.
［22］Geiger A, Lenz P, Stiller C, et al. Vision meets robotics: The KITTI dataset ［J］. [WTBX][STBX]International Journal of Robotics Research[STBZ][WTBZ], 2013, 32(11): 1231-1237.
［23］Geiger A. Vision meets robotics: The KITTI dataset［J］. [WTBX][STBX]International Journal of Robotics Research[STBZ][WTBZ], 2013, 32 (11): 1231-1237.