Astronomical Spectra Denoising Based on Simplified SURE-LET Wavelet Thresholding

The aim of astronomical spectra denoising is to reduce the noise level, while preserving astrophysical information hidden in spectra. Many astronomical spectra denoising approaches exist, in which wavelet thresholding approaches are widely adopted in many literatures. For spectra denoising using wavelet thresholding, the thresholding function greatly influences the quality of the resulting denoised spectra. In this paper, a new orthogonal wavelet-based denoising approach is proposed, which is a composite approach of the simplified SURE-LET thresholding and the soft thresholding. Experimental results demonstrate that higher signal-to-noise ratio (SNR) and better visual quality can be obtained by the proposed approach.