An Improved Decision based Asymmetric Trimmed Median Filter for Removal of High Density Salt and Pepper Noise

Abstract

improved decision based asymmetric trimmed median filter (IDBATMF) algorithm for the restoration of gray scale, and color images that are highly corrupted by salt and pepper noise (SPN) is proposed in this paper. The proposed algorithm selects a 3 × 3window around pixel being processed. If the selected window consists of 0's, 255's and other random values, then the noisy pixel is replaced with the trimmed median value of the elements in selected window. When all the pixel values are 0's and 255's then the decision is taken based on numbers of 0 and 255 values in the window. If 6 or more pixels are 0 (or 255) then the pixel being processed is replaced with 0 (or 255), else it is replaced with mean of value of all the elements present in the selected window. Setting the pixel as 0 or 255 ensures that a large black or white patch of the image is restored satisfactorily. The proposed algorithm shows better results than existing median filters including the recent Modified Decision Based Unsymmetrical Trimmed Median Filter (MDBUTMF). Experimental results show improvements both visually and quantitatively compared to that of the MDBUTMF and other standard filters.