Nonlinear regularization techniques for seismic tomography

DOI10.1016/J.JCP.2009.10.020MaRDI QIDQ2655679zbMATH OpenOpenAlexFDO

Authors Ignace Loris, H. Douma, Guust Nolet, Ingrid Daubechies, C. Regone

Publication date 25 January 2010

Published in Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0808.3472

inverse problem one-norm regularization sparsity tomography wavelets

Nontrigonometric harmonic analysis involving wavelets and other special systems (42C40) Seismology (including tsunami modeling), earthquakes (86A15) Inverse problems in geophysics (86A22)

Abstract: The effects of several nonlinear regularization techniques are discussed in the framework of 3D seismic tomography. Traditional, linear,

e l l_{2}

penalties are compared to so-called sparsity promoting

e l l_{1}

and

e l l_{0}

penalties, and a total variation penalty. Which of these algorithms is judged optimal depends on the specific requirements of the scientific experiment. If the correct reproduction of model amplitudes is important, classical damping towards a smooth model using an

e l l_{2}

norm works almost as well as minimizing the total variation but is much more efficient. If gradients (edges of anomalies) should be resolved with a minimum of distortion, we prefer

e l l_{1}

damping of Daubechies-4 wavelet coefficients. It has the additional advantage of yielding a noiseless reconstruction, contrary to simple

e l l_{2}

minimization (`Tikhonov regularization') which should be avoided. In some of our examples, the

e l l_{0}

method produced notable artifacts. In addition we show how nonlinear

e l l_{1}

methods for finding sparse models can be competitive in speed with the widely used

e l l_{2}

methods, certainly under noisy conditions, so that there is no need to shun

e l l_{1}

penalizations.

Recommendations

Cites work

Cited in

(14)

Describes a project that uses

Uses Software

This page was built for publication: Nonlinear regularization techniques for seismic tomography

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2655679)