Using Anfis with circular polygons for impulsive noise suppression from highly distorted images

In this paper, a novel approach is presented to the restoration of images corrupted by impulsive noise (IN), with a new nonlinear IN suppression filter, entitled circular polygons based adaptive-fuzzy filter (CF). The proposed filter is based on statistical impulse detection and nonlinear filtering which uses adaptive-network-based fuzzy inference system (Anfis) as a missed data interpolant over the circular polygons and provides estimates for the original intensity values of corrupted pixels. Impulse detection is realized by using the chi-square based goodness-of-fit test, which yields a decision about the impulsivity of each pixel. Extensive simulations were realized to demonstrate the capability of CF and they reveal that the proposed filter achieves a better performance than the other filters mentioned in this paper in the cases of being effective in noise suppression and detail preservation, also when the images are highly corrupted by IN.