On the Estimation of Wavelet Coeecients
نویسنده
چکیده
In wavelet representations, the magnitude of the wavelet coeecients depends on both the smoothness of the represented function f and on the wavelet. We investigate the extreme values of wavelet coeecients for the standard function spaces A k = ff j kf (k) k 2 1g, k 2 N. In particular, we compare two important families of wavelets in this respect, the orthonormal Daubechies wavelets and the semiorthogonal spline wavelets. Deriving the precise asymptotic values in both cases, we show that the spline constants are considerably smaller.
منابع مشابه
Incorporating Information on Neighboring Coeecients into Wavelet Estimation
In standard wavelet methods, the empirical wavelet coeecients are thresholded term by term, on the basis of their individual magnitudes. Information on other coeecients has no innuence on the treatment of particular coeecients. We propose a wavelet shrinkage method that incorporates information on neighboring coeecients into the decision making. The coeecients are considered in overlapping bloc...
متن کاملBayesian Inference with Wavelets: Density Estimation
We propose a prior probability model in the wavelet coeecient space. The proposed model implements wavelet coeecient thresholding by full posterior inference in a coherent probability model. We introduce a prior probability model with mixture priors for the wavelet coeecients. The prior includes a positive prior probability mass at zero which leads to a posteriori threshold-ing and generally to...
متن کاملBayes Wavelet Shrinkage
Wavelet shrinkage is a novel method for data denoising and function estimation. M uller and Vidakovic (1995) propose a hierarchical prior on the wavelet coeecients and shrink them by applying the induced Bayes rule. In this paper, a diierent and more elastic hierarchical prior is elicited on the model parameters describing the wavelet coeecients. Exact Bayesian analysis is impossible and the sh...
متن کاملContextual Hidden Markov Models forWavelet - domain Signal
Wavelet-domain Hidden Markov Models (HMMs) provide a powerful new approach for statistical model-ing and processing of wavelet coeecients. In addition to characterizing the statistics of individual wavelet coeecients, HMMs capture some of the key interactions between wavelet coeecients. However, as HMMs model an increasing number of wavelet coeecient interactions , HMM-based signal processing b...
متن کاملAdaptive Thresholding of Wavelet
Wavelet techniques have become an attractive and eecient tool in function estimation. Given noisy data, its discrete wavelet transform is an estimator of the wavelet coeecients. It has been shown by Donoho and Johnstone (1994) that thresholding the estimated coeecients and then reconstructing an estimated function reduces the expected risk close to the possible minimum. They ooered a global thr...
متن کامل