Denoising by Averaging Reconstructed Images: Using Singularity Function Analysis

A newfound method of denoising that based onAveraging Reconstructed Image (AVREC), is used. Theapproach was proposed on signals, approximately about lastdecade, Since 2004. In definition (procedure), first of all, wedivide the spectrum of noisy image into several images that canbe then, reconstructed with 2-D Singularity Function Analysis(SFA) model. Among this mathematical model, each matrix or adiscrete set of data, represents as a weighted sum of singularityfunctions. In image denoising field, this technique, rebuilt all losthigh frequencies parameters that are essential. Illustrate eachnew image, as a sum of noise-free image and the small noise. Soon, we can then, denoise image by averaging reconstructed ones.Both theoretical and experimental results on standard gray-scaleimages, confirm the advantages (benefits) of this approach as anapplicable method of denoising.

Keywords:

Image Processing , denoising , reconstruction , SFA(Singularity Function Analysis)