An adaptive parameterized block-based singular value decomposition for image de-noising and compression

摘要

We propose an adaptive parametrized block-based singular value decomposition (APBSVD) for preserving the edge structure and avoiding blurred image after the compressing process. We also exploit a proper selection of the Peak Signal to Noise Ratio (PSNR) parameter at each block to appropriately eliminate high-frequency noise. Our numerical results show that the proposed methods are promising which effectively remove the noise, and preserves the edge information well during the block-based SVD image compression.