An efficient denoising approach for removal of impulse noise

In this paper, an algorithm is presented for images which are highly corrupted by salt and pepper noise also called Impulse Noise. The proposed filtering algorithm consist of two phases. First phase identify the noisy pixels from corrupted image and then in second phase a filter is used to reconstruct the noisy pixel. The experimental result shows that the proposed filtering algorithm performs better than simple median filter for removing the noise and preserves the edges. Since the algorithm is simple and effective, it is suitable for many real time applications.