An Efficient Image Denoising Based on Weiner Filter and NeighSure Shrink

Abstract

Weiner filter denoise the image using linear stochastic framework. It eliminates the noise by estimating optimal filter for noisy input image by minimizing the mean square error between the desired image and estimated image. The main drawback of this filter is the performance is reduced when the noise is random and unknown as it has fixed frequency response for all frequencies. The efficiency of this filter can be increased by incorporating method noise thresholding using NeighSure shrink. This paper presents a method which is a blend of Weiner filter and wavelet based NeighSure shrink thresholding. The results indicates that the proposed method is significantly superior than wavelet thresholding, Weiner filter and Gaussian filter with its method noise thresholding techniques in terms of visual quality, Peak Signal to noise ratio and image quality index.