全极化频谱外推的合成孔径雷达成像分辨率增强方法

全极化合成孔径雷达可对不同极化通道分别独立进行频谱外推来增强分辨率,但分别独立处理难以利用极化信息的冗余性与互补性．将目标全极化数据分解到反映三种独立散射机理的Pauli基上,对分解后的全极化数据进行三维联合频谱外推处理,并利用实测数据进行了对比分析．结果表明：极化分解后,通过三维联合频谱外推的方法相比传统方法具有更高的预测模型匹配精度,可获得更好的分辨率增强性能．

Enhanced resolution in fully polarimetric synthetic aperture radar imaging using bandwidth extrapolation

The resolution of polarimetric SAR (PolSAR) can be improved by using the conventional method to different polarization-channel data independently. However, it doesn’t exploit the polarization information entirely, which essentially presents the target scattering. In this paper, full-polarization data is decomposed according to Pauli’s basis. Under this decomposition, we develop a PolSAR bandwidth extrapolation algorithm, which is jointly extrapolating the frequency bandwidth of all polarization-channel data. By the joint extrapolation, the proposed method can obtain more accurate prediction model than the conventional bandwidth extrapolation method, yielding higher resolution. Experimental results of real data are presented to demonstrate the effectiveness of the proposed method.