摘要
针对提高舰船目标SAR图像分辨率的问题,基于Fourier字典的稀疏表示法得到的结果中杂波和旁瓣较多,正则化方法对于点目标密集的图像处理结果较差。提出一种稀疏表示法与正则化方法相结合的超分辨率方法,首先利用基于Fourier字典的稀疏表示法对原始数据进行处理,然后建立所得结果的正则化模型,并利用正则化方法消除杂波、旁瓣及噪声。理论分析及仿真计算结果表明,此方法可以较好地解决上述两种方法存在的问题,得到较好的超分辨率结果 。
There are two common methods to improve the SAR image resolution of ship target. The result of sparse representa- tion method based on Fourier dictionary contains a lot of sidelobe and clutter, and the result of regularization method is not good for the image processing of dense points. This paper proposes a superresolution algorithm which combines the sparse representation method and regularization method. Firstly, the sparse representation method based on Fourier dictionary is used to process the raw data. Then, the regularization model is established. And the regularization method is used to eliminate the cutter, sidelobe and noise. The results of theoretical analysis and simulation show that the proposed algorithm can solve the problems of the two algo- rithms mentioned above and can get good superresolution results.
出处
《遥测遥控》
2015年第3期15-22,共8页
Journal of Telemetry,Tracking and Command
