An improved compromise for soft/hard thresholds in wavelet denoising
Huizhong Yang · Caai Transactions on Intelligent Systems · 2008
The characteristics of two traditional wavelet compromise methods,referred to as the soft-thresholding method and the hard-thresholding method,were analyzed in the wavelet domain.A new compromise thresholding method that improves performance is presented,based on the standard compromise threshold method,yet more flexible and easier to treat mathematically.Compared with traditional wavelet compromise methods,the new approach avoids the discontinuity of the hard-threshold method and also decreases the fixed bias between the estimated wavelet coefficients and the decomposed wavelet coefficients of the soft-threshold method.Simulation results show that the improved method is better at denoising than the soft-threshold method,the hard-threshold method,the standard compromised threshold method,as well as other threshold function methods.