An Enhanced Two-Stage Impulse Noise Removal Technique based on Fast ANFIS and Fuzzy Decision

Image enhancement plays a vital role in various applications. There are many techniques to remove the noise from the image and produce the clear visual of the image. Moreover, there are several filters and image smoothing techniques available in the literature. All these available techniques have certain limitations. Recently, neural networks are found to be a very efficient tool for image enhancement. A novel two-stage noise removal technique for image enhancement and noise removal is proposed in this paper. In noise removal stage, Adaptive Neuro-Fuzzy Inference System (ANFIS) with a Modified Levenberg-Marquardt training algorithm was used to eliminate the impulse noise. The usage of Modified LevenbergMarquardt training algorithm will reduce the execution time. In the image enhancement stage, the fuzzy decision rules inspired by the Human Visual System (HVS) are used to categorize the image pixels into human perception sensitive class and nonsensitive class, and to enhance the quality of the image. The Hyper trapezoidal fuzzy membership function is used in the proposed technique. In order to improve the sensitive regions with higher visual quality, a Neural Network (NN) is proposed. The experiment is conducted with standard image. It is observed from the experimental result that the proposed FANFIS shows significant performance when compared to existing methods.