A New Recursive Structure-Adaptive Filter for Impulse Noise Removal

摘要

A New Recursive Structure-Adaptive Filter for effective suppression of impulse noise is presented in this paper. In the first stage, at every noisy location in a corrupted image the structure activity is identified by quadtree decomposition. Based on the local structure activity and Noise statistics, a recursive window adaptive filtering mechanism is proposed to switch between simple mean, median and center weighted median filters to provide detail preserving filtering of corrupted pixels. Extensive simulations show that the proposed filter restores fairly well even the images that are highly corrupted.