全局一致的多视图像配准

为了提高多视图像配准的精度和一致性，考虑到各种不确定信息，提出了一个基于概率估计的多视图像配准方法。把各视点的位置姿态作为系统状态，构造系统增广模型和系统观测模型，运用增广卡尔曼滤波技术对各视点的位置姿态进行递归地估计，实现了精确的、全局一致的多视图像配准。实验对提出的配准方法进行了验证，结果表明提出的多视图像配准方法能够有效地处理不确定信息，配准结果是一致的和精确的。

Globally consistent solution to multiview image registration

In order to improve the accuracy and consistency of image registration,an approach for multiview image registration based on probability estimation is proposed by taking account of various uncertainties.The position and orientation of all viewpoints are considered as system state.System augmented model and system observation model are constructed.The position and orientation of viewpoints are estimated recursively with augmented Kalman filter.Multiview image registration with high accuracy and consistency is realized.Some experimental results are provided to validate the performance of the proposed method.The proposed multiview image registration method can handle the uncertainty efficiently.Registration result is accurate and consistent.