Adaptive Threshold Selection Using Principle of SURE (The irreductible variance term is not included, it does not change the position of the minimum).
SURE_MSEthresh(wcn, wcf, tresh, diagWWt, b, sigma, hatsigma, policy,
keepwc = TRUE)Noisy wavelet coefficents.
True wavelet coefficients.
Threshold values.
Weights.
Thresholding type (b=1: soft, b=2: JS).
Sd of the noise.
Estimator of the sd (if any).
Dependent or uniform.
Boolean allowing to export the coefficients of the frame after thresholding (TRUE by default).
res a dataframe contening MSE, SURE, hatSURE and their respective minima
Note: - the calculation of the MSE is also included for comparison purpose.
de Loynes, B., Navarro, F., Olivier, B. (2021). Data-driven thresholding in denoising with Spectral Graph Wavelet Transform. Journal of Computational and Applied Mathematics, Vol. 389.