.png){kind=icon}
.png){kind=icon}
Return
Prestack Diffraction Separation Using DTW and f-x SVMD
.png){kind=icon}
PRE
AI
DOI：10.1109/LGRS.2025.3629674.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Diffractions indicate the presence of small-scale subsurface geological discontinuities. However, strong reflections often obscure the high-resolution information carried by diffractions, making it essential to suppress reflections before performing diffraction imaging. We propose a shot-domain diffraction separation method combining successive variational mode decomposition (SVMD) with an improved dynamic time warping (DTW). Conventional shot-domain separation methods exploit seismic velocity to enhance kinematic differences between reflected and diffracted waves, rendering their effectiveness sensitive to the accuracy of the velocity model. In contrast, the DTW method automatically flattens seismic events based on their dynamic characteristics, eliminating the need for seismic velocity. We improve the DTW method by incorporating regularization techniques and neighborhood distance computation, allowing it to flatten reflections in shot gathers without flattening diffractions. By applying the improved DTW method, we transform shot gathers into flattened reflection-extended gathers, where reflections exhibit low-dip characteristics and diffractions present high-dip features. Leveraging this distinction, we develop a dip-angle filter based on the SVMD method to separate diffractions in flattened reflection-extended gathers. The numerical experiment confirms that the proposed method is capable of detecting 5 m karst caves and faults with a 2 m throw. The field data example demonstrates that the proposed method can effectively remove reflections while preserving diffractions, thereby facilitating high-resolution imaging of the geological discontinuities characterized by diffractions.
Keywords:
Diffraction separation
dynamic time warping (DTW)
flattened reflection-extended gathers
successive variational mode decomposition (SVMD)
Journal
I
IF:
4.4
Papers:
579
Citations:
0