多摄像机系统位姿估计的广义正交迭代算法

目前, 多摄像机系统的位姿估计还缺乏系统的方法, 它通常通过求解透视n点问题或者求解使两组3D点集之间平方和误差最小的刚体变换来解决, 这些方法都有局限性。正交迭代算法是基于点特征的单目视觉算法, 快速且全局收敛, 是目前性能最优的实时位姿估计算法之一, 被广泛应用。提出了一种广义正交迭代算法把所有摄像机获取的全部图像作为整体计算得到相对位姿参数, 是通用的多目视觉位姿估计算法。算法先把所有摄像机数据进行统一表达, 再把所有摄像机观测到的全部特征点的目标空间共线性误差平方和作为误差函数, 最后经数学推导得到使该误差函数最小化的迭代求解过程。实验结果验证了算法的有效性以及在多摄像机系统位姿估计中的优越性。

Abstract

It still lacks of systematic method for solving pose estimation of a multiple camera system (MCS). The pose-estimation problem is usually achieved by solving the Perspective-n-point problem or finding the least-squared-error rigid transformation between two 3D point sets. They have limitations. The orthogonal iteration (OI) algorithm is charalteristic can converge fast and globally is based on points. And it is one of the state-of-art real-time pose estimation algorithms and used widely. A generalized iterative algorithm is proposed. It is a universal multi-view algorithm for pose estimation. In the method, all images acquired from the cameras are treated as a whole, and the relative pose parameter is calculated this method, datum of multiple cameras is unified, and the sum of object-space collinearity error of feature points all cameras sensed is used as error function for MCS. An iterative algorithm is developed to minimize the error function. The experiment result demonstrates its effectiveness and superiority of the algorithm for pose estimation of MCS.

参考文献

[1] . 航天器交会计算机视觉系统测距求解新算法[J]. 北京航空航天大学学报, 2006, 7(32): 764-768.

. New algorithm of solving for ranges during final approach of spacecraft rendezvous[J]. J. Beijing University of Aeronautics and Astronautics, 2006, 7(32): 764-768.

[2] . Fischler, R. C. Bolles. Random sample consensus: A paradigm for model fitting and automatic cartography[J]. Comm. ACM, 1981, 6(24): 381-395.

[3] . G. Lowe. Fitting parameterized three-dimensional models to images[J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 1991, 13(5): 441-450.

[4] . P. Lu, G. Hager, E. Mjolsness. Fast and globally convergent pose estimation from video images[J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 2000, 22(5): 610-622.

[5] . Linear pose estimation from points or lines[J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 2003, 25(5): 578-589.

[6] . Pose estimation for augmented reality applications using genetic algorithm[J]. IEEE Trans. on System, Man, and Cybernetics, 2005, 36(6): 1295-1301.

[7] . Robust pose estimation from a planar target[J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 2006, 28(12): 2024-2030.

[8] . 航天器交会对接位姿测量迭代算法[J]. 哈尔滨工业大学学报, 2005, 37(8): 691-694.

. An iterative method for vision-based relative pose parameters of RVD spacecrafts[J]. J. Harbin Institute of Technology, 2005, 37(8): 691-694.

[9] . 一种基于视觉成像的快速收敛的位姿测量算法及实验研究[J]. 航空学报, 2007, 28(4): 943-947.

. A fast convergent pose estimation algorithm and experiments based on vision images[J]. Acta Aeronautica et Astronautica Sinica, 2007, 28(4): 943-947.

[10] Mario L. Fravolini, Lorenzo Pollini, Brian Stolarik. A comparison of pose estimation algorithms for machine vision based aerial refueling for UAVs[C]. Proceedings of the 2006 Conference on Control and Automation, Mediterranean, June 2006. 1～6

[11] C. Fermüller, Y. Aloimonos, P. Baker et al.. Multi-camera networks: eyes from eyes[C]. Proc. IEEE Workshop on Omnidirectional Vision, USA, 2000. 11～18

[12] . Davis. Constraint integration for efficient multiview pose estimation with self-occlusions[J]. IEEE Trans on Pattern Analysis and Machine intelligence, 2008, 30(3): 493-506.

[13] Fredrik Viksten, Robert S. Oderberg, Klas Nordberg. Increasing pose estimation performance using multi-cue integration[C]. Proceedings of the 2006 IEEE International Conference on Robotics and Automation, Orlando, Florida , May 2006. 3760～3767

[14] . K. P. Horn. Closed-form solution of absolute orientation using unit quaternion[J]. J. Opt. Soc. Am., 1987, 4: 629-642.

[15] . S. Arun, T. S. Huang, S. D. Blostein. Least-squares fitting of two 3-D point sets[J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 1987, 9(5): 698-100.

许允喜, 蒋云良, 陈方. 多摄像机系统位姿估计的广义正交迭代算法[J]. 光学学报, 2009, 29(1): m00006. Xu Yunxi, Jiang Yunliang, Chen Fang. Generalized Orthogonal Iterative Algorithm for Pose Estimation of Multiple Camera Systems[J]. Acta Optica Sinica, 2009, 29(1): m00006.

多摄像机系统位姿估计的广义正交迭代算法

关于本站 Cookie 的使用提示

全站搜索

多摄像机系统位姿估计的广义正交迭代算法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索