多极化SAR图像3D-SPIHT压缩

该文针对多极化合成孔径雷达(SAR)图像在极化通道之间的相关性，提出了3D-SPIHT压缩方法。将多极化SAR图像(HH，HV，VV图像)作为一个整体，进行三维矩阵变换。首先在极化通道之间进行一维DCT变换，极化平面内进行二维离散小波变换(DWT)，然后对3个极化混合系数平面采用分级树的集合划分(SPIHT)算法进行嵌入式统一混合编码。由于不是单独处理每一极化图像，因此不仅可以去除各极化图像内部之间的相关性，也可以去除极化通道之间的相关性。另外，由于采用统一嵌入编码，码流具有完全嵌入性，可以实现3个极化平面之间比特的精确自动分配。理论推导和仿真结果表明该方法对多极化SAR图像压缩是十分有效的。

3D-SPIHT Compression of Multi-polarimetric SAR Images

This paper proposes a method for multi-polarization Synthetic Aperture Radar (SAR) images compression: 3D-SPIHT，which considers the multi-polarimetric SAR images(HH,HV,VV) as a 3D-matrix unit, then does 3D-matrix transform to it. First, 1D-DCT(Discrete Cosine Transform) is processed among the polarimetric channel and then 2D-DWT(Discrete Wavelet Transform) is processed in the polarimetric SAR image plane, finally, 3D-SPIHT algorithm is adopted to encode the mixed coefficients of three frames entirely. This method does not process every channel image separately, so it not only can remove the redundancies of the image inside, but also can remove the redundancies among the polarimetric channels. And whats more, the embedded code stream can implement the precise bit allocation among the polarization channels. Both the theory and the experimental results show that the method in this paper is efficient.