Robust outliers exclusion for high density impulse noise suppression algorithm

Abstract

An algorithm for denoising images polluted by high impulse noise density rate is proposed in this paper. The proposed algorithm is specifically designed to stand against burst of impulses. In such a case, the test window is filled with impulses due to its initial limited size(typically 3×3). As a result, the standard median filter fails completely in recovering corrupted pixel. To remedy this effect, larger test window size is used to increase the probability of removing the effect of impulse bursts. Thereafter, all impulses are removed from the test window, and a trimmed mean with 25% of outlier exclusion for the rest noise free pixels is used to replace the corrupted pixel. The experimental results show that the performance of the proposed algorithm is superior to that of the conventional state-of-the-art denoising approach.