An efficient method for blocky and scratch noise removal

In this paper, we propose a new method which can remove blocky noises and scratch noises efficiently. Our method employs the idea of robust principal component analysis (RPCA) and uses similar patches from natural images as much as possible. In our method, a matrix is formed whose columns are the patches, which are then decomposed into a noise matrix and a noiseless matrix. Experimental results show that our method can achieve excellent results when compared to state-of-the-art methods.