A Salt and Pepper Noise Removal and Restoration Refinement Algorithm

Noise removal is an important task inside the image processing area. In this paper, an algorithm for reducing salt and pepper noise and improving the restoration quality through refinement is presented. This algorithm proposes a double screen. As a first step, the algorithm computes an estimation of the denoised image by using an adaptive median filter. Then, the Non-Local Means algorithm is used in order to have a better quality of reconstitution. Obtained results show that the implementation of this proposal gives considerable noise suppression, even with high noise densities, and it could prepare the image for other processes like image segmentation and object recognition.