基于改进SVA的SAR/INSAR图像处理

针对现有的Spatially Variant Apodization (SVA)算法不能有效抑制旁瓣或损失主瓣能量的问题,本文提出了一种改进的SVA算法.该算法把传统的滤波器从3点扩展到5点,并且根据采样率的不同,设定相应的滤波器参数,得到满足约束优化理论的最优解.该算法适用于任意奈奎斯特采样率,既能有效地抑制旁瓣,又能保持主瓣的能量和信号的高分辨率;同时能够在一定程度上提高图像的信噪比,在干涉操作中增强复图像对的相关性.实验结果表明,与传统的频域加窗方法相比,该方法能够在保持图像高分辨率的前提下,更有效地抑制旁瓣;同时提高干涉操作的精度.

SAR/INSAR image processing based on modified spatially variant apodization

The existing Spatially Variant Apodizations (SVAs) either cannot depress the sidelobes effectively or reduces the energy of the mainlobe. To improve this, the modified SVA (MSVA) of this paper, which expands the traditional filter from 3-tap to 5-tap and sets relevant parameters according to different sampling rates, can get the excellent result that satisfies constrained optimization theory. This method is suitable for any Nyquist sampling rate, and can both depress the sidelobes effectively and keep the energy of the mainlobe and the resolution of the image; at the same time, this method can improve the SNR of the image partly and enhance the coherence of image pair. The method can reduce sidelobe levels more effectively than classical amplitude weighting, while maintaining the image resolution, and improve the accuracy of the interferometric operation, which is demonstrated by the result of the experiment.