Noise reduction based on fuzzy logic

Digital images consist of noises for various reasons. Due to presence of these noises, obtaining the necessary information is difficult or impossible.In this study, fuzzy logic rules are used for image processing. According to the concept of classical set: an element belong to the cluster or the element is not belong to the cluster; but according to the concept of fuzzy set, all of the elements having a degrees of membership at the range of [0,1] belong to the cluster. This situation, provide to produce the result using methods closer to human reasoning system.Several fuzzy filtering methods for noise reduction have been developed so far. In this study, aimed that salt and pepper noise is reduced using a novel filtering method while preserving edge, detail and texture information. Filtering process is realized in two stages. In the first stage, noise level is detected in the image, in the second stage; filtering process is done with Switching Adaptive Fuzzy Filter (SAFF).The proposed method is able to reduce salt and pepper noise in gray level digital images while preserving image details, and textures very well.All results of the applications were obtained by using the MATLAB program.