Based on multi-resolution analysis of wavelet, this article is aimed at building a new soft threshold function for wavelet de-noising, to overcome the discontinuous disadvantage of the hard threshold function. In the area adjacent to the threshold, continuously adjustable nonlinear functions are introduced in piecewise to process the wavelet coefficients more carefully in this area. Consequently, large deviation caused by super-compression of wavelet coefficients when using traditional soft threshold could be avoided, and the nonlinearity of the system is able to be effectively kept. Using both signal to noise ratio (SNR) and mean square error (MSE) as the evaluation indicators, simulation results show that the improved method is more effective than the method based on traditional hard and soft threshold.
[1]
G..
A Theory for Multiresolution Signal Decomposition : The Wavelet Representation
,
2004
.
[2]
David L. Donoho,et al.
De-noising by soft-thresholding
,
1995,
IEEE Trans. Inf. Theory.
[3]
Stéphane Mallat,et al.
A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
,
1989,
IEEE Trans. Pattern Anal. Mach. Intell..
[4]
Shi Zhi-wei,et al.
Chaotic Signal Denoising Based on Threshold Selection of Wavelet Transform
,
2005
.
[5]
Martin Vetterli,et al.
Adaptive wavelet thresholding for image denoising and compression
,
2000,
IEEE Trans. Image Process..