二维EMD 方法在微尺度波斜率波数谱提取中的应用

进行海面微尺度波频率分析时，为消除重力波影响，引入二维 EMD(经验模态分解)方法滤除造成频率混叠的长重力波，然后进行傅立叶变换，得到海面微尺度波谱，进一步处理得到斜率波数谱。斜率波数谱提取结果表明：采用二维EMD方法可以有效滤除图像中混叠的长波信息，在此基础上进行微尺度斜率波谱计算的结果与国外已有结果基本一致。文中还将二维 EMD 方法与小波方法的处理结果进行了比对，结果表明在滤除长波的影响时，二维EMD方法具有其较明显的优势。

The computation method of slope wave-number spectrum of micro-scale waves based on the two-dimensional empirical mode decomposition

To eliminate the effects of long gravity waves in spectral analysis, we used two-dimensional empirical mode decomposition (2D-EMD) to strain off long gravity waves. Then we conducted Fourier transform to get sea surface micro-scale spectrum. After further processing, the slope of the wave-number spectrum can be computed. The results showed that 2D-EMD method can effectively filter out long-wave aliasing information in images. The results of micro-scale slope wave-number spectrum were consistent with foreign. We compared the results between 2D-EMD and wavelet method. The results showed that 2D-EMD method had significant advantages for long-wave filter.