基于Gabor小波变换的机载SAR海面风向反演方法  

A Method of Ocean Surface Wind Direction Retrievals for Airborne SAR Images Based on Gabor Wavelet Transform

作　　者：张日伟[1] 严卫[1] 艾未华[1] 马烁[1] 

机构地区：[1]解放军理工大学气象学院,南京211101

出　　处：《微波学报》2011年第5期79-83,共5页Journal of Microwaves

基　　金：国家自然科学基金(41005018)

摘　　要：给出了一种基于Gabor小波变换的机载合成孔径雷达(SAR)海面风向反演的新方法。该方法利用Gabor小波对SAR图像进行二次小波分解,并对小波系数作FFT变换来获取图像谱,其低波数谱连线的垂线方向就是海面风场的风向。利用该方法获得了SAR图像海面风向信息,并与àtrous算法反演结果、浮标测得的海面风向(真值)进行了比较。实验结果表明,采用该方法获得的机载SAR海面风向反演结果与海面浮标实测数据吻合,比àtrous算法有较大改进。A new method of ocean surface wind field retrieval for airborne SAR images based on Gabor wavelet transform is proposed in this paper.A multi-resolution decomposition has been applied to the SAR image by using the Gabor wavelet in this method,and the two-dimensional FFT is applied to the wavelet coefficient to get the image spectrum.The vertical direction of the connecting line for the low wave spectrum is the direction of the ocean surface wind field.The ocean surface wind direction is retrieved by this method and the result is compared with that retrieved from the à trous algorithm and measured from the buoy.It is shown that the retrieval of the ocean surface wind direction by Gabor wavelet is in good agreement with the true value measured by the buoy,and is better than à trous algorithm.

关 键 词：合成孔径雷达 海面风向 小波变换 àtrous算法 GABOR小波 

分 类 号：TN959.73[电子电信—信号与信息处理]