Noise Suppression from images is one of the most important concern in digital image processing. Impulse noise is one such noise, which may corrupt images during their acquisition or transmission or storage. Several algorithms are used for removing salt and pepper noise (SPN) from corrupted images. The proposed algorithm is to remove high density impulse noise using modified unsymmetric trimmed root mean square (RMS) filter. The pixel values other than 0 and 255 are left unaltered. Only the noisy pixels (with the values 0 or 255) are replaced by the RMS value of the remaining pixel elements of the selected window. The effect of denoising in an images are analyzed using different parameters like peak signal to noise ratio (PSNR), and mean square error (MSE).
Volume 11 | 04-Special Issue
Pages: 1205-1210