工业增强现实中的相机跟踪

潘绍松; 左洪福

doi:10.3788/OPE.20111906.1353

您当前的位置：

首页 >

文章列表页 >

工业增强现实中的相机跟踪

信息科学 | 更新时间：2020-08-12

- 工业增强现实中的相机跟踪
- Camera tracking in industrial augmented reality
- 光学精密工程 2011年19卷第6期页码：1353-1359
- 作者机构：
  
  南京航空航天大学民航学院,江苏南京,210016
- 作者简介：
- 基金信息：
  
  国家自然科学基金联合基金资助项目(No.60939003/F01)()
- DOI：10.3788/OPE.20111906.1353
  中图分类号： TP391.1
- 收稿日期：2010-06-29，
  
  修回日期：2010-10-26，
  
  网络出版日期：2011-06-25，
  
  纸质出版日期：2011-06-25
- 稿件说明：
移动端阅览
潘绍松, 左洪福. 工业增强现实中的相机跟踪[J]. 光学精密工程, 2011,19(6): 1353-1359

PAN Shao-song, ZUO Hong-fu. Camera tracking in industrial augmented reality[J]. Editorial Office of Optics and Precision Engineering, 2011,19(6): 1353-1359
潘绍松, 左洪福. 工业增强现实中的相机跟踪[J]. 光学精密工程, 2011,19(6): 1353-1359 DOI： 10.3788/OPE.20111906.1353.

PAN Shao-song, ZUO Hong-fu. Camera tracking in industrial augmented reality[J]. Editorial Office of Optics and Precision Engineering, 2011,19(6): 1353-1359 DOI： 10.3788/OPE.20111906.1353.

摘要

提出了利用工业对象表面共面圆进行相机位姿跟踪的解决方案。首先

利用共面圆构建对象坐标系;用扩展卡尔曼滤波器预测相机位姿限定椭圆轮廓点搜索的位置及范围

计算梯度归一化互相关系数来形成轮廓点候选数据集;然后

剔除候选数据集中的野值

用最大似然估计法进行高精度椭圆拟合;最后

根据二次曲线的射影理论计算相机位姿并更新滤波器的状态。实验表明

本文方法跟踪相机位姿的最大误差角度为1.4

平移为3.5 mm

跟踪速度为10~12 frame/s

满足工业增强现实中相机位姿跟踪的应用需求。

Abstract

A solution for camera tracking based on coplanar circles from industrial objects was presented. Firstly

a coordinate system for an object frame was constructed based on coplanar circles. Then

the Extended Kalman Filter(EKF) was exploited to predict camera pose to restrict the positions and regions for searching ellipse contours

and the candidate contour points of ellipse were formed by calculating normalized cross coefficient of gradient. With maximum likelihood estimation

a high accurate ellipse was obtained after eliminating outliers from candidate dataset. Finally

the camera pose was estimated using projective theory of conics

and the filter state was updated. Experiments show that the maximum pose deviation of camera tracking is 1.4 for rotation and 3.5 mm for translation

and the tracking rate is 10~12 frame/s.The results can meet the requirement of pose tracking in industrial Augmented Reality(AR).

关键词

Keywords

references

王涌天,林精敦. 随机树特征匹配算子性能研究[J]. 北京理工大学学报,2009,29(11): 988-993. WANG Y T, LIN J D. A study on the feature matching performance of randomized trees [J]. Transactions of Beijing Institute of Technology, 2009,29(11):988-993. (in Chinese)[2] 李利军,刘伟,等. 基于粒子滤波的虚实配准方法研究[J]. 工程图学学报,2009(1):119-124. LI L J, LIU W. A new registration method based on particle filter [J]. Journal of Engineering Graphics, 2009(1):119-124.(in Chinese)[3] BLESER G,PASTARMOV Y. Real-time 3D camera tracking for industrial augmented reality applications [J]. Journal of WSCG, 2005:47-54.[4] DHOME M, RICHETIN M, LAPRESTE J T. Determination of the attitude of 3D objects from a single perspective view [J]. IEEE Trans. Pattern Anal. Mach. Intell., 1989,11(12):1265-1278.[5] ANSAR A, DANIILIDIS K. Linear pose estimation from points or lines [J]. IEEE Trans. Pattern Anal. Mach. Intell., 2003,25(5):578-589.[6] PHILIP DAVID, DANIEL DEMENTHON. SoftPOSIT: simultaneous pose and correspondence determination [J]. Intl. Journal of CV, 2004,59(3):259-284.[7] CHEN Q, WU H Y. Camera calibration with two arbitrary coplanar circles . Proc. European Conf. CV, 2004:521-532.[8] WANG G,WU Q. Pose estimation from circle or parallel lines in a single image [J]. Lecture Notes in Computer Science, 2007,4844:363-372.[9] GURDJOS P. Euclidean Structure from N2 parallel circles: theory and algorithms . Proceedings of the 9th European Conference on Computer Vision, Graz, Austria, 2006(1):238-252.[10] 潘绍松,左洪福. 利用共面圆特征实现增强现实图像注册[J]. 光电工程,2010(8):81-85. PAN SH S, ZUO H F. Registration in augmented reality from coplanar circles [J]. Opto-Electronic Engineering, 2010(8):81-85.(in Chinese)[11] KAEWAPICHAI W,KAEWTRAKULPONG P. Robust ellipse detection by fitting randomly selected edge patches . World Academy of Science, Engineering and Technology,2008,48:30-34.[12] THANH M N, SIDDHANT A. A real-time ellipse detection based on edge grouping . Proceedings of the 2009 IEEE international conference on Systems, Man and Cybernetics, 2009:3280-3286.[13] 赵其杰,屠大维. 基于Kalman滤波的视觉预测目标跟踪及其应用[J]. 光学精密工程, 2008,16(5):937-942. ZHAO Q J,TU D W. Kalman filter based vision predicting and object tracking method and its application [J]. Opt. Precision Eng., 2008,16(5):937-942.(in Chinese)[14] 孙宏伟,房建成. 椭圆拟合方法在磁罗盘罗差校准中的应用[J]. 光学精密工程, 2008,17(12):3034-3039. SUN H W,FANG J CH. Application of ellipse fitting method to calibration of magnetic compass deviation [J]. Opt. Precision Eng., 2008,17(12):3034-3039.(in Chinese)[15] HARTLEY R,ZISSERMAN A. Multiple View geometry in Computer Vision [M].Second Edition. London: Cambridge University Press, 2003.[16] BOGDAN M P M. Reduction of bias in maximum likelihood ellipse fitting . Proc. Int. Conf. Patt. Recog., 2000(13):3802-3086.[17] KANATANI. Compact algorithm for strictly ML ellipse fitting . Proc. Int. Conf. Patt. Recog., 2008, Tampa, FL,2008:1-4.[18] KENICHI, PRASANNA. Hyperaccurate ellipse fitting without iterations [J]. Memoirs of the Faculty of Engineering, 2010(44):42-49.[19] VINCENT E, LAGANIERE R. Detecting planar homographies in an image pair . Proceedings of the 2nd International Symposium on Image and Signal Processing and Analysis, Pula , Croatia, 2001:182-187.[20] OFIR P,WERMAN M. A linear time histogram metric for improved SIFT matching . Proceedings of the 10th European Conference on Computer Vision, Berlin, Heidelberg, 2008,5304:495-508.

浏览量

244

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

LightDiffu-DCE：基于光照强度扩散的低光照图像增强

多级图特征融合引导相机位姿回归

基于机载的红外动态目标视频实时超分辨率重建

基于知识蒸馏的Transformer视觉跟踪器

3D-CNN与Transformer混合结构的高光谱图像空谱联合分类