摘要： 提出了一种快速易行的时间序列图像自动配准算法：首先对两幅图像分别进行离散小波变换，然后对分解后的两幅低频图像计算联合直方图，最后采用改进的联合直方图区域计数方法来设计目标函数并进行配准参数优化。实验结果与对比分析表明：在基本不影响图像配准精度的情况下，算法大大加快了图像配准速度。

关键词: 时间序列图像；图像配准；小波变换；联合直方图

Abstract: Serial images refer to an aggregate of images of the same scene captured in serial time. This paper proposes a new method to solve the problem of serial images registration. The source images are decomposed by the wavelet transform. The lowest frequency bands of the source images are used to estimate the registration parameters. The traditional joint histogram based registration algorithm is improved by applying new region-counting method. Experiments show that the proposed algorithm can increase the speed of registration

Key words: Serial images; Image registration; Wavelet transform; Joint histogram