中文摘要：

为了实现快速、高精度的自动图像配准，提出了基于遗传算法和小波分解的图像配准方法。此方法在小波分解的基础上，以互信息为相似性测度利用遗传算法为搜索算法对图像进行一次配准，然后以特定步长在小范围内对配准参数进行修正，对图像进行二次配准。该方法实验结果表明该方法具有较快的速度且显著的改善了配准精度。

英文摘要：

To improve the efficiency and accuracy in image registration, a method for the image registration based on genetic algorithm and wavelet decomposition is proposed. Rigid transformation is supposed between the floating image and the template image. Decomposed wavelet approximate components of the original image are used to improve the efficiency in image registration. The mutual information is used as the individual fitness evaluation function of the genetic algorithm. As the soul of the algorithm, the transformation parameters between two images can be obtained. To improve the accuracy, the transformation parameters obtained are used as the starting value in the search. Better parameters are then searched in a small range with a fine step. The example shows that the method has the advantages of efficiency and accuracy.