An improved median filter for removing extensive salt and pepper noise

In this paper, an improved median filtering algorithm is proposed for the restoration of images that are highly corrupted by salt and pepper noise. The proposed algorithm (PA) replaces the noisy pixel with the median of the already processed pixel values of the selected window, in which all the pixels are noisy, by considering two complementary processing directions. It turns out that PA can efficiently remove the salt and pepper noise when the noise level is higher than 60%. Simulation results are given to illustrate that PA can be applied to not only gray scale but also color image denoising and outperforms the existing methods in terms of the qualitative and quantitative measures of the restored image.