Signal and image denoising via wavelet thresholding: Orthogonal and biorthogonal, scalar and multiple wavelet transforms (Q2712151)

The aim of this substantial paper is to investigate the wavelet thresholding in the context of scalar orthogonal, scalar biorthogonal, multiple orthogonal, and multiple biorthogonal wavelet transforms. The covariance structure of the transform is studied for both scalar and vector types of multiwave thresholding, and the careful formulation of the universal threshold represents the key to the excellent results obtained in the extensive numerical simulations reported within the paper for signal and image denoising. For both the 1D and 2D cases, the reconstructed signals derived from the proposed general method demonstrate much reduced noise levels; an average noise reduction gain of 7.5dB was achieved over test examples.NEWLINENEWLINEFor the entire collection see [Zbl 0958.00020].