Remote sensing image reconstruction based on Shearlet transform and total generalized variation regularization

摘要

This paper, we propose a method that exploit total generalized variation (TGV) and Shearlet transform for image restoring. The TGV adaptive regularize different image regions at different levels and the Shearlet transform can efficiently represent image anisotropic features such as edges, curves. A new image restoration model combining TGV and Shearlet transform is proposed for image restoration. The proposed model is solved by splitting variables and applying the alternating direction method of multiplier (ADMM). Experimental results show that the proposed algorithm can effectively restore image and improve the quality.