Image denoising based on wavelet transform

Abstract

Several wavelet denoising methods based on wavelet transform, such as modulus maximum and threshold denoising, have been proposed in this paper. In modulus maximum method, we use a new piecewise cubic spline interpolating (PCSI) algorithm instead of alternate projection (AP) method to reconstruct wavelet coefficients. Middle threshold is presented based on wavelet shrinkage. It overcomes not only the discontinuous of hard-threshold algorithm but also the shortcoming that there is an invariable dispersion between the estimated wavelet coefficients and decomposed wavelet coefficients of soft-threshold algorithm. At the same time, it is more elastic than the soft- threshold and hard-threshold function.