Dip Moveout and Stolt Migration Using the Singular Value Decomposition
, by Stewart A. Levin and Fabio Rocca
Phase shift migration, after simple modifications,
produces accurate images of synthetic aperture array radar (SAR) signals.
Approximating the matrix of phase shifts with the largest term of
its singular value decomposition (SVD) yields an analog of Stolt migration.
This SVD-Stolt migration permits rapid computation, often without undue
loss of accuracy.
In this paper, we apply the SVD method to seismic processing
by developing a rapid method for dip moveout (DMO) in constant velocity.
We also use SVD to study extensions to Stolt migration
when velocity is no longer constant with depth.
Lastly, we suggest several further applications and directions for
future research.