Statistical Moments and Fuzzy Logic Based Classification of Noise Present in Digital Images

Abstract: In this paper we proposed, a method which effectively classifies noise present in the images using statistical moment based feature extraction and continuing with fuzzy based classification. Noise in corrupted information that may hide the original information of an image. To identify which type of noise exactly present in the digital images fuzzy logic method is used in this work. The feature value helps us to distinguish between the noises. In order to remove the noise from the images one should know which type of noise present in that image. As the step of identifying of noise completes it makes easy for the researcher to remove the noise present in the image Gaussian noise, Salt and pepper noise, Speckle noise have an different feature value which is generated by skewness and kurtosis and also are the type of noises that are identified in this paper using Fuzzy Classification.

Keywords: kurtosis, skewness, Gaussian noise, Salt and pepper noise, Speckle noise, Fuzzy classification.