Salt-and-Pepper Noise Suppression with A Novel Two Stage Algorithm

摘要

In this paper, we propose a new algorithm to remove the salt-and-pepper noise from the corrupted images. At first, we employ an efficient impulse noise detector to locate the position of the noisy pixels and store the information. Secondly, a detail-preserving variational filter based on ℓ~1 norm is used to remove these noise. The experiments compare the proposed algorithm with the standard median filter and other two stage denoising algorithms from both quality and quantity. The results show that our algorithm have the better visual effect and peak signal noise ratio.