A new Adaptive Decision based Robust Statistics Estimation Filter for high density impulse noise in images and videos

In this paper an adaptive decision based robust statistics estimation filter to remove high density impulse noise is presented. The proposed algorithm detects the corrupted pixel at the initial stage and replaces them with an estimated image data using the proposed robust estimation algorithm. The new robust statistics based filter shows significantly better image quality than standard median filter, weighted median filter, recursive weighted median filter, adaptive median filter, decision based algorithm and the recently proposed robust estimation algorithm. The proposed method removes the noise efficiently even at noise level as high as 90% and preserves the edges. The proposed filter has better image quality then existing nonlinear filters.