Classification-based weighted filter for image corrupted by impulse noise

A novel filter for image corrupted by impulse noise is presented. In this approach, each current pixel under consideration will be replaced by its expected value. For that, local neighboring pixels are used to provide a measure of similarity according to the property of impulse noise, by which each pixel of the filter window is classified into either noise-free region or corrupted region. In order to avoid the influence of impulse noise on the estimation of the expected value, only the pixels of noise-free region are used to estimate the expected value based on distance weighs. Simulation results indicate that the proposed filter impressively outperforms other techniques in terms of noise suppression and detail preservation across a wide range of impulse noise corruption, ranging from 1% to 90%, especially where there is a high probability of impulse noise