On image denoising with wavelets - A case study by considering a priori distortion

The problem of image denoising based on wavelets is considered. The paper uses an image denoising method by imposing a distortion input parameter instead of threshold. The method has two steps. The first step builds a dependency, linear or nonlinear, between the final desired quality (PSNR) and the necessary parameter to select the details coefficients. The second algorithm step performs denoising based on the parameter computed on the previous step. The threshold level is computed by estimating the probability density function (PDF) of the details coefficients and having the probability of the coefficients which must be kept. Roughly, the obtained results are at better quality levels than other well known denoising methods.