Every digital image has a two-dimensional mathematical representation of the digital image. Digital image are made out of pixels i.e. picture component. Every pixel speaks to the dark level for highly contrasting photographs at a solitary point in the image, so a pixel can be spoken to by a small speck of particular shading. Image restoration is the process of restoring degraded images which cannot be taken again or the process of obtaining the image again is costlier. We can restore the images by prior knowledge of the noise or the disturbance that causes the degradation of the image. Image restoration is done in two domains: spatial domain and frequency domain. In the spatial domain, the filtering action for restoring the images is done by directly operating on the pixels of the digital image. In our research work different format of the same image will be executed for a different level of noise and then we will analyze which format will be best and besides PSNR two more parameters MSE and IEF also considered. In our research work, our main objective is to remove salt and pepper noise from the image. As in base paper, 30% and 70% salt and pepper noise are removed with PSNR value. But in our dissertation work salt and pepper noise at 30%, 50%, 70%, and 75% are removing with three parameters like PSNR, MSE, and IEF. After the filtering, the image is remapped into spatial domain by inverse Fourier transform to obtain the restored image. Different noise models were studied. Different filtering techniques in both spatial and frequency domains were studied and improved algorithms were written and simulated using Matlab. Restoration efficiency was checked by taking peak signal to noise ratio (PSNR) and mean square error (MSE) into considerations.