Research on the decision-based adaptive weighted mean filter algorithm for impulse noise removal

The decision-based adaptive weighted mean filter is proposed to remove impulse noise from the highly corrupted image. The proposed filter first identifies the corrupted pixels using the soft decision-based noise detector and then removes the detected impulses using the adaptive weighted mean filter while keeping the uncorrupted pixels unaltered. Extensive simulations indicate that the proposed filter significantly outperforms a number of existing decision-based filters in that it can remove impulse noise from the corrupted image effectively while preserving the details in the image very well.