Wavelet Denoising with New Double Thresholds Function
LU Shize
LI Xingye
Abstract：The traditional hard threshold function have the disadvantage of discontinuity ,which tends to produce Gibbs oscillation phenomenon .And the soft threshold function have a fixed deviation problem when it process the wavelet coeffi‐cients .And the threshold determined by the sqtwolog rule is significantly large ,which leads to losing the high frequency part of signal .A new double thresholds function is constructed .The double thresholds function is continue in the thresholds ,and the wavelet coefficients processed by the double thresholds function not have a fixed deviation problem .It is a good way to keep the high frequency part of signal by adjusting the double thresholds .The simulation results show that the new double thresholds function dealing with noised signal is more effective to inhibit Gibbs oscillation phenomenon and signal distortion . And it improves signal‐to‐noise ratio significantly and reduces the mean square error .
Keywords：waveletsignal-to-noise ratiothreshold functiondenoisingdouble thresholds
Publication Date：2015-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 1634-1639 )