Images are often corrupted by noise due to errors generated in noisy sensors or communication channels. It is important to eliminate noise in the images before some subsequent processing. In this paper it is proposed to obtain the denoised estimate in spatial domain method using filters like mean filter, Gaussian filter, Weiner filter, median filter, midpoint filter, unsharp filter and progressive switching median filter and combination of these filters. Form the observations of PSNR for various filters, it is inferred that progressive switching median filter is suitable for denoising salt and pepper noise. These methods were simple and easy to apply but their effectiveness is limited since this often leads blur or smoothed out in high frequency regions. New and better approaches perform thresholding in wavelet domain of an image. The idea of wavelet thresholding relies on the assumption that the signal magnitude dominates the magnitude of the noise in wavelet representations, so that wavelet coefficients can be set to zero if their magnitudes are less than a predetermined threshold. In this paper it is proposed that VISU shrink is effective because it is not subband adaptive.

Key words: PSNR, progressive switching median filter, VISU shrink, SURE shrink, Bayes shrink