Salt and pepper noise removal with image inpainting

Salt and pepper noise removal is an important task in image processing. In this paper, we propose a simple and efficient restoration algorithm with the theory of image inpainting. Our algorithm takes noisy pixels as missing data for inpainting, adaptively selects convolution mask in terms of details of local regions, and achieves restoration by iterative convolutions. Many experiments are conducted to validate efficiency of our algorithm. The results show that our algorithm can efficiently remove noise while preserving image details even at a high noise density, and outperforms some well-known algorithms in terms of peak signal-to-noise ratio.