Performance of the decision based algorithm for the removal of unequal probability salt and pepper noise in images

A Decision based unsymmetrical trimmed variants is proposed for the elimination of unequal probability salt and pepper noise in gray scale images. The proposed algorithm replaces the corrupted pixels with unsymmetrical trimmed median or midpoint based on the noisy pixels in the current processing window. The proposed algorithm shows excellent noise elimination capability with good detail preserving nature. The proposed algorithm exhibits high PSNR, SSEVI.