基于加窗的相位相关的多谱段图像匹配

Multispectral image matching based on windowed phase correlation

图像匹配是计算视觉领域的基本问题,多谱段图像匹配一直是图像匹配中的一个难点,被广泛应用于土地覆盖、变化检测分析、图像融合中。在利用FFT实现相位相关的基础上,针对其存在频谱泄露和边缘效应的问题,提出一种加窗预处理的相位相关法,对合成孔径雷达图像和光学图像进行匹配,并通过核回归方法生成二维窗函数。实验结果表明基于加窗的相位相关对多谱段图像匹配有效。

Image matching is a basic problem in the field of computational vision. Multispectral image matching has always been a difficult point in image matching. It is widely used in land cover, change detection analysis, and image fusion. Aimed at the problem of spectral leakage and edge effect, on the basis of using FFT to achieve phase correlation, a phase correlation method with windowing preprocessing is proposed to match the synthetic aperture radar image and the optical image and generate a two-dimensional window function through the kernel regression method. The experimental results show that the phase correlation based on windowing is effective for multispectral image matching.