L- and Θ-curve approaches for the selection of regularization parameter in geophysical diffraction tomography

Since inverse problems are usually ill-posed it is necessary to use some method to reduce their deficiencies. The method that we choose is the regularization by derivative matrices. When a first derivative matrix is used the order is called the first. Then, second-order regularization is when the ma...

Full description

Saved in:
Bibliographic Details
Published inComputers & geosciences Vol. 33; no. 5; pp. 618 - 629
Main Authors Santos, E.T.F., Bassrei, A.
Format Journal Article
LanguageEnglish
Published Oxford Elsevier Ltd 01.05.2007
Elsevier Science
Subjects
Applied geophysics
Diffraction tomography
Earth sciences
Earth, ocean, space
Exact sciences and technology
Hydrocarbons
Internal geophysics
Inverse problems
L-curve
Regularization
Regularization parameter
Sedimentary rocks
Θ-curve
Inverse problems
L-curve
Θ-curve
Regularization
Regularization parameter
Diffraction tomography
reservoir rocks
inverse problem
vertical seismic profiles
diffraction
hydrocarbons
data processing
digital simulation
velocity
tomography
geometry
geophysical methods
computers
Online AccessGet full text

Cover

More Information
Summary:Since inverse problems are usually ill-posed it is necessary to use some method to reduce their deficiencies. The method that we choose is the regularization by derivative matrices. When a first derivative matrix is used the order is called the first. Then, second-order regularization is when the matrix is formed by second-order differences, and order zero means that the regularization matrix is the identity. There is a crucial problem in regularization, which is the selection of the regularization parameter λ. We used the L-curve as a tool for the selection of λ, and we propose a new extension, which we call the Θ-curve. The tool was applied in geophysical diffraction tomography in two acquisition geometries: cross-hole and vertical seismic profile (VSP), where the goal is to obtain the 2-D velocity distribution from the measured values of the scattered acoustic field. We present several simulation results with synthetic data, for the three regularization orders mentioned above. We validate the necessity of some kind of regularization, as well as the feasibility of both parameter selection approaches.
ISSN:0098-3004
1873-7803
DOI:10.1016/j.cageo.2006.08.013