AI based automated identification and estimation of noise in digital images

Abstract

Noise identification, estimation and denoising are the important and essential stages in image processing techniques. In this paper, we proposed an automated system for noise identification and estimation technique by adopting the Artificial intelligence techniques such as Probabilistic Neural Network (PNN) and Fuzzy logic concepts. PNN concepts are used for identifying and classifying the images, which are affected by the different type of noises by extracting the statistical features of noises, PNN performance is evaluated for classification accuracies. Fuzzy logic concepts such as Fuzzy C-Means clustering techniques have been employed for estimating the noise affected to the image are compared with the other existing estimation techniques.