Enhanced Modified Decision-Based Unsymmetric Trimmed Adaptive Neighborhood Winsorized Mean Filter for Removing 1–99% Levels of Salt-and-Pepper Noise

Abstract

In this paper, a novel and effective algorithm for removing 1–99% levels of salt-and-pepper noise is proposed. The proposed algorithm comprises a two-phase scheme. The first phase involves a combination of enhanced modified decision-based unsymmetric trimmed mean filter (EMDBUTMF) and decision-based unsymmetrical trimmed modified winsorized mean filter (DBUTWMF), and the second phase is employed after the first phase to replace the left over noisy pixels. The proposed algorithm is the fusion of the benefits of EMDBUTMF, DBUTWMF and decision-based diagonal neighborhood pixel algorithm and shows better results than the decision-based algorithm, modified decision-based unsymmetric trimmed median filter (MDBUTMF), EMDBUTMF, Modified decision-based unsymmetrical trimmed median filter with trimmed global mean (MDBUTMF_GM), DBUTWMF. The proposed algorithm is examined against 1–99% levels of salt-and-pepper noise for several grayscale bitmap images, and it gives better peak-signal-to-noise ratio, image enhancement factor and Structural Similarity Index Measures values for high noise densities.