Abstract
Abstract in Undetermined
We introduce a new method for prestack depth migration of seismic common-shot gathers.
The computational procedure follows standard steps of the reverse-time migration, i.e.,
downward continuation of the `source' and the `receiver' waveelds, followed by application
of an imaging condition (e.g. zero-lag cross-correlation of these elds). In our method we
rst nd a sparse data representation with a small number of Gaussian wave packets. We
then approximate the downward waveeld propagation (for the `source' and the `receiver'
elds) by a `rigid'
ow of these wave packets along seismic rays. In this case, the wave
packets are simply translated and rotated according to the ray geometry. One advantage
with using Gaussian wave packets is that analytic formulas can be used for translation, rotation, and the application of the cross-correlation imaging condition. Moreover, they
allow more sparse representations than competing methods. In combination, this yields a
computationally and memory ecient migration procedure, as only few rays have to be
traced, and since it is cheap to compute the cross-correlation for the intersecting rays.
We introduce a new method for prestack depth migration of seismic common-shot gathers.
The computational procedure follows standard steps of the reverse-time migration, i.e.,
downward continuation of the `source' and the `receiver' waveelds, followed by application
of an imaging condition (e.g. zero-lag cross-correlation of these elds). In our method we
rst nd a sparse data representation with a small number of Gaussian wave packets. We
then approximate the downward waveeld propagation (for the `source' and the `receiver'
elds) by a `rigid'
ow of these wave packets along seismic rays. In this case, the wave
packets are simply translated and rotated according to the ray geometry. One advantage
with using Gaussian wave packets is that analytic formulas can be used for translation, rotation, and the application of the cross-correlation imaging condition. Moreover, they
allow more sparse representations than competing methods. In combination, this yields a
computationally and memory ecient migration procedure, as only few rays have to be
traced, and since it is cheap to compute the cross-correlation for the intersecting rays.
Original language | English |
---|---|
Pages (from-to) | 83-106 |
Journal | Studia Geophysica et Geodaetica |
Volume | 56 |
Issue number | 1 |
DOIs | |
Publication status | Published - 2012 |
Subject classification (UKÄ)
- Computer Vision and Robotics (Autonomous Systems)
- Mathematics
Free keywords
- seismic imaging
- sparse representation
- Gaussian wave packets
- migration
- beam