Adaptive Rank-ordered Switching Median Filter for Salt & Pepper Impulse Noise Reduction

Abstract

An adaptive rank-order based switching median filter is proposed in this paper for the restoration of digital images corrupted with salt & pepper impulse noise. A more ideal and efficient impulse noise filter, the ARSMF detects all the corrupted pixels even when the image is highly corrupted and restores them with a more perfect signal detected from an appropriate region whose dimensions vary adaptively with noise levels. This noise replacing signal does the dual functions of noise suppression and detail preservation than many other variants of impulse noise filters. Experimental results, the subjective and objective analysis add to the efficiency and fidelity claim of the proposed ARSMF