Novel robust RM-KNN filters with impulsive noise suppression for image processing

Abstract

We introduce novel robust filtering algorithms applicable to image processing. They are derived by use of RM-type point estimations and the restriction technique of the well-known, specifically for image processing, KNN filter. The derived RM-KNN filters effectively remove impulsive noise while edge and fine details are preserved. The proposed filters are tested on simulated images and real imaging radar data and provide excellent visual quality of the processed images and good quantitative quality in comparison with the standard median filter. Recommendations to obtain the best processing results by proper selection of derived filter parameters are given.