基于运动轨迹的单目相机位姿自标定方法

详细信息    查看全文 | 推荐本文 |

英文篇名：Trajectory based monocular camera pose self-calibration method 作者：张墨逸 ; 张秋余 ; 段宏湘 ; 杨树强 英文作者：ZHANG Moyi;ZHANG Qiuyu;DUAN Hongxiang;YANG Shuqiang;School of Computer and Communication,Lanzhou University of Technology; 关键词：相机标定 ; 单目视觉 ; 与视角无关 ; 行为识别 ; 多视角几何 ; 校正 英文关键词：camera calibration;;monocular vision;;view-invariant;;action recognition;;multiple-view geometry;;rectification 中文刊名：HZLG 英文刊名：Journal of Huazhong University of Science and Technology(Natural Science Edition) 机构：兰州理工大学计算机与通信学院; 出版日期：2019-02-15 19:05 出版单位：华中科技大学学报(自然科学版) 年：2019 期：v.47;No.434 基金：国家自然科学基金资助项目(61363078);; 兰州理工大学模式识别国家重点实验室开放课题基金资助项目(201700005) 语种：中文; 页：HZLG201902012 页数：6 CN：02 ISSN：42-1658/N 分类号：69-74

摘要

为了减小轨迹图像分析中的透视效应,使用提取出的运动平面目标特征点,基于相机小孔成像模型与目标特征点运动约束,构建了一种相机位姿的自标定模型来标定相机位姿.通过标定的相机位姿,可将轨迹进行重投影,实现了将轨迹图像校正为正投影图像.实验结果表明:本标定模型使用跟踪到的运动目标进行相机标定,相机位姿的标定精度优于张正友标定法.
        In order to reduce the perspective effect in trajectory analysis,using the object's motion points abstracted,a camera self-calibration model was constructed to compute the pose of the camera.Then,the trajectory was re-projected to obtain the orthographic projected trajectory which was view-invariant.Experimental results demonstrated that the proposed model is more accurate than ZHANG Zhengyou's calibration method in pose calibration.This model used the tracked moving target for camera calibration.This proposed model can be used widely in camera pose calibration and view-invariant trajectory analysis.

引文

[1]LEE T,KIM C,CHO D D.A monocular vision sensor-based efficient SLAM method for indoor service robots[J].IEEE Transactions on Industrial Electronics,2019,66(1):318-328.
    [2]AVIDAN S,SHASHUA A.Trajectory triangulation:3Dreconstruction of moving points from a monocular image sequence[J].Pattern Analysis and Machine Intelligence IEEE Transactions on,2000,22(4):348-357.
    [3]LI B,DAI Y,HE M,et al.A relaxation method to articulated trajectory reconstruction from monocular image sequence[C]//Proc of IEEE China Summit and International Conference on Signal and Information Processing.Xi’an:IEEE,2014:389-393.
    [4]SHIRATORI T,MATTHEWS I,SHEIKH Y Aet al.3Dreconstruction of a moving point from a series of 2Dprojections[C]//Proc of Computer Vision-ECCV 2010.Berlin:Springer,2010:158-171.
    [5]TSAI R Y.A versatile camera calibration technique for high-accuracy 3D machine vision metrology using offthe-shelf TV cameras and lenses[J].IEEE Journal on Robotics and automation,1987,3(4):323-344.
    [6]ZHANG Z Y.A flexible new technique for camera calibration[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334.
    [7]STEPHEN T,DANAIL S,CRISPIN S,et al.Hand-eye calibration for rigid laparoscopes using an invariant point[J].International Journal of Computer Assisted Radiology and Surgery,2016,11(6):1071-1080.
    [8]ZHANG K,FANG Z,LIU J,et al.Fusion of vision and IMU to track the racket trajectory in real time[C]//Proc of IEEE International Conference on Mechatronics and Automation.Takamatsu:IEEE,2017:1769-1774.
    [9]XU Y,YANG S,MA J,et al.Simultaneous calibration phantom commission and geometry calibration in cone beam CT[J].Physics in Medicine and Biology,2017,62(17):375-390.
    [10]TAKAHASHI K,NOBUHARA S,MATSUYAMA T.Mirror-based camera pose estimation using an orthogonality constraint[J].IPSJ Transactions on Computer Vision and Applications,2016,8:11-19.
    [11]LIU T,XU W,YIN X,et al.Planar metric rectification from perspective image via vertical vanishing point[C]//Proc of International Congress on Image and Signal Processing.Dalian:IEEE,2015:951-957.
    [12]TANG H,LIU Y.Automatic simultaneous extrinsicodometric calibration for camera-odometry system[J].IEEE Sensors Journal,2018,18(1):348-355.
    [13]ZHANG M Y,ZHANG Q Y,DUAN H X,et al.Self-calibration based view-invariant hand gesture trajectory analysis[J].Journal of Information Hiding and Multi-media Signal Processing,2018,9(5):1114-1127.
    [14]张墨逸,张秋余,芦俊池,等.基于结构特征点的手势建模与高层特征提取[J].华中科技大学学报:自然科学版,2015,43(10):87-91.