为更好地利用合成孔径雷达SAR图像本身的风条纹信息来反演高精度的海面风向，采用不同模式、不同极化方式的星载ASAR图像，基于抽样小波变换结合快速傅里叶变换的方法反演海面风向，同时将该方法反演的结果与传统的傅里叶变换法和改进的局部梯度法反演的结果进行了对比。结果表明，抽样小波变换应用于海面风向的反演是可行的，反演的风向精度优于传统的傅里叶变换法和改进的局部梯度法。ASAR不同模式、不同极化方式的数据反演结果存在着一定的差异。

To make better use of wind streak information from Synthetic Aperture Radar (SAR) images used in turn to obtain high-precision sea surface wind fields, wind direction is retrieved using decimated wavelet transform associated with two-dimensional Fast Fourier Transform (FFT) with different modes and Advanced Synthetic Aperture Radar (ASAR) images of various polarizations. Finally, the resulting wind direction is compared with those obtained using traditional FFT and the improved local gradient method. The results demonstrate the feasibility of applying the wavelet transform algorithm to sea surface wind direction inversion. Multi-resolution analysis of ASAR images with time-frequency window of wavelet transform enables more precise determination of wind direction compared with traditional Fast Fourier transform and improved local gradient method. In addition, the inversion results contain certain differences in diverse modes and various polarization data.