Removing of high density salt and pepper noise using fuzzy median filter

Abstract

This paper presents the removal of high density salt and pepper noise in gray scale Images using fuzzy based median filter (FBMF) algorithm. FBMF replaces the noisy pixel by median value when 0's, 255's and other pixel values are present in the chosen window and when all pixel values are either 0's or 255's, or combination of these, then the noise pixel is replaced by fuzzy membership value of a selected window. FBMF algorithm is tested on different images and compared with the existing Median Filter (MF), adaptive median filter (AMF), Decision Based Algorithm (DBA), Modified decision based unsymmetrical trimmed median filter algorithm (MDBUTMF). Results shows that FBMF gives better Peak Signal to Noise Ratio (PSNR) and Image Enhancement Factor (IEF) compared with the existing algorithms in the literature.