Wavelet Transform Based On Image Denoising Using Thresholding Techniques

Abstract: wavelet transforms enable us to represent signals with a high degree of scarcity .This is the principle behind a non-linear wavelet based signal estimation technique known as wavelet denoising. wavelet thresholding is a signal estimation technique that exploits the capabilities of wavelet transform for signal denoising. The aim of this project was to study various techniques such as visuShrink, SureShrink, NeighShrink(proposed method) and determine the best one for image denoising. VisuShrink and SureShrink, the thresholding application removes the coefficients that are in significant to some threshold. NeighShrink is an efficient image denoising algorithm based on the decimated wavelet transform (DWT). Its disadvantage is to use a suboptimal universal threshold and identical neighbouring window size in all wavelet subbands. In this paper, an improved method is given, which can determine an optimal threshold and neighbouring window size for every subband by the Stein’s unbiased risk estimate (SURE). In NeighShrink, optimal threshold and neighborhood window size in all subbands keep unchanged. In NeighShrink (proposed method), the Optimal threshold and Neighborhood window size in all subbands is changed. In NeighShrink(proposed method) we retain the required information from the removed coefficients by using neighborhood window size and optimal threshold. They threshold the wavelet coefficients in overlapping blocks rather than individually or term by term as VisuShrink or SureShrink.

Keywords: Image denoising, MSE, PSNR, Wavelet transforms, Neighborhood.