Title :
Regularization by multiscale models
Author :
Fabre, Eric ; Benveniste, Albert
Author_Institution :
IRISA, Rennes, France
Abstract :
We describe multiscale stochastic models, tightly linked to the wavelet transform, and the associated estimation algorithms. This framework can be used to model stochastic fractals or more generally to handle signals that “live” at different interacting resolutions. We then show how these models can be used as prior information to regularize geological data stemming from a nuclear sensor that is itself multiresolution
Keywords :
Markov processes; array signal processing; estimation theory; fractals; geology; geophysical signal processing; signal resolution; stochastic processes; wavelet transforms; Gauss-Markov models; estimation algorithms; geological data regularization; multiresolution sensor; multiscale signal processing; multiscale stochastic models; nuclear sensor; signal resolutions; stochastic fractals; wavelet transform; Band pass filters; Equations; Filter bank; Filtering; Fractals; Geology; Low pass filters; Signal resolution; Stochastic processes; Wavelet transforms;
Conference_Titel :
Digital Signal Processing Workshop, 1994., 1994 Sixth IEEE
Conference_Location :
Yosemite National Park, CA
Print_ISBN :
0-7803-1948-6
DOI :
10.1109/DSP.1994.379869
