Azimuth Ambiguity Suppression for Single Look Complex Image Based on Block Sparse Representation

Azimuth ambiguity affects image quality seriously, which poses a great challenge to post-processing methods. However, existing ambiguity suppression method mainly deals with the azimuth ambiguity in raw data or utilizing multi-channel information. This paper proposes a blind adaptive method for single-look complex (SLC) images with less prior information. The energy of azimuth ambiguity will be sparsely spread in each sub-look image by multi-look processing on the SLC image. Through robust principal component analysis, useful information embedded in each sub-look image is effectively reconstructed and the azimuth ambiguity is removed. By applying sub-look synthesis, the problem of azimuth resolution loss caused by sub-look images is effectively avoided. Experimental results demonstrate the effectiveness of the proposed method.