RESTORATION OF IMAGES CORRUPTED BY HIGH DENSITY SALT & PEPPER NOISE THROUGH ADAPTIVE DECISION BASED UNSYMMETRICAL TRIMMED MEAN FILTER

Abstract

An adaptive decision based unsymmetrical trimmed mean filter algorithm for the restoration
of the images that are corrupted by high density salt and pepper noise is suggested in this paper. The
proposed algorithm first checks for noise and noise free pixel in the selected window, if the pixel value lies
between 0 to 255 considered as noise free pixel else noisy pixel. The noise free pixels are left unaltered. The
noisy pixels are replaced by adaptive decision unsymmetric based mean filter. The proposed algorithm
removes the salt and pepper noise at high noise density and preserves the edges and fine details of the image.
The proposed algorithm shows better results than the SMF,AMF,PSMF,DBA, MDBUTMF algorithms. The
proposed algorithm is verified with different standard images and shows better PSNR, and IEF values when
compared to the other filtering algorithms.

Author

R.Vinoth Kumar , Dr.R.Sudhakar