Entropy-based estimation of salt-pepper noise in wavelet domain

In this paper, we proposed an algorithm for estimating the density of salt & pepper noise in images with entropy inspection in wavelet domain. Based on the trait that energies of image signal and noise could be separated by wavelet transform, and on the fact that noise entropy in wavelet domain changes with approximate logarithm mode along with the noise level, we exhibit how the entropy values of noisy images in wavelet domain change with the noise level in quantitative form, and indicate that such relation is robust to individual images. We thus take advantage of this relation to make estimation of noise level. Simulation results demonstrate that the proposed algorithm is better than the present methods while providing more robust and more exact data.