Mixed Noise Removal Using Adaptive Median Based Non-Local Rank Minimization

In this paper, we present an innovative mechanism for image restoration problems in which the is corrupted by a mixture of additive white Gaussian noise (AWGN) and impulse (IN). Mixed removal much more challenging problem contrast to where either only one type model (either or impulse) involved. Several well-known efficient algorithms exist effectively remove Impulse noise, independently. However, practice, may occur as such models. Thus, existing techniques devised handle individual types not perform well. Moreover, complexity hinges on fact that from given affects statistics residual image. Therefore, rigorous required infers altered but also removes effective manner. regard, approach introduced restore underlying three key steps. Firstly, intensity values, affected impulsive are identified analyzing with help adaptive median filtering. The values then aggregated exploiting nonlocal data redundancy prior. Thus first step enables remaining follow zero mean distribution filtered Secondly, estimate resulting image, acts parameter subsequent singular value thresholding process rank minimization. Finally, reduced optimization applied pre-processed obtained step. experimental results indicate proposed AMNLRA (Adaptive Median based Non-local Low Rank Approximation) can tackle mixed compared numerous state art algorithms.