Complex Compatible-Structure Tensor Total Variation Regularization for High Resolution SAR Imaging

Conventional sparsity-driven Synthetic Aperture Radar (SAR) imagery often encounters the difficulty of extraction for complicated structural features. To this end, this paper proposes a Complex Compatible-Structure Tensor Total Variation (CC-STV) optimization algorithm, which can describes and enhances the structural feature of SAR imagery. By the introduction of multi-channel structure tensor, the CC-STV optimization algorithm can utilize detailedly local neighborhood information to describe and enhance structural features of SAR imagery when sparse features enhancement suppresses detailed structural feature. Considering the enhancement of structural features in sparsity-driven SAR images, Alternating Direction Multi-Multiplier (ADMM) algorithm is introduced to realize cooperative multi-feature enhancement. In the experiments, efficiency and superiority are validated by SAR simulated and raw data. Compared with conventions, the superiority of the proposed algorithm in terms of quantitative aspect is proven by the analysis experiment of phase transition.