Additive white Gaussian noise level estimation based on block SVD

Accurate estimation of noise level is of fundamental interest in a wide variety of vision and image processing applications as it is critical to the processing techniques that follow. In this paper, a new, effective noise level estimation method is proposed based on existing noise estimation methods in SVD domain. The analysis and experiments results demonstrate that the proposed algorithm can reliably infer noise levels, and in comparison with the existing method in SVD domain, the proposed algorithm can reduce the computation complexity greatly.