Noise suppression using block-based singular value decomposition filtering

Noise suppression is one of the most essential processes for image processing. The goal of the noise suppression is how to remove the noise while keeping the important image features as much as possible. In this paper, we present an adaptive block-based singular value decomposition method for noise suppression. Instead of applying block-based singular value decomposition (BSVD) directly to noisy images, we suggest to apply BSVD on the edge with noise image version obtained from the difference between the original noisy image and its blur version. From the experiments, the objective and subjective test results show that our proposed approach compared with conventionally methods can suppress noise, preserve the important image features as well as effectively smooth in the smooth region.