提出一种新颖快速的亚像素级大尺寸平移图像配准算法以实现大尺寸高精度的图像配准。首先利用下采样互相关函数解混叠模型实现像素级的粗定位, 降低了互相关矩阵的傅里叶变换维度, 并减小了离散傅里叶变换矩阵的乘法数目以加快粗配准过程; 然后在粗定位点邻域范围内利用改进的矩阵乘法对离散傅里叶变换(DFT)高倍数重采样, 并采用双向搜索策略进行亚像素快速定位。仿真实验结果表明, 相比常规的基于FFT算法的图像配准, 本文提出的算法将大大减少配准的时间复杂度和空间复杂度, 而不会损失配准精度。

In order to achieve large-scale and high-precision image registration, a novel and fast sub-pixel large-scale translation image registration algorithm was proposed. Firstly, the coarse positioning at the pixel level was achieved by using the down-sampling cross-correlation aliasing model, which reduced the Fourier transform dimension of the cross-correlation matrix and the multiplication of the discrete Fourier transform matrix, so as to speed up the coarse registration process. Then, the improved DFT multiplier of the matrix multiplication was used in the neighborhood of the coarse point, and the sub-pixel fast location was achieved by the bidirectional search strategy. Simulation experiment results show that compared with common image registration based FFT algorithm, the proposed algorithm could greatly reduce space complexity and time complexity without losing accuracy.