Seismic diffractions separation and imaging based on convolutional neural network

Abstract

In exploration seismology, the reflections have been extensively used for imaging and inversion to detect hydrocarbon and mine resources, which are generated from subsurface continuous impedance interfaces. When the reflectors are not continuous and their size reduces to less than a half wavelength, the reflected wave becomes scattering, which is also known as the diffraction. Both reflection and diffraction can be used to image subsurface structures, and the latter is helpful to resolve small-scale discontinuities, such as fault plane, pinch-out, Karst caves and salt edge. However, the amplitudes of diffractions are usually much weaker than that of reflections. This makes it difficult to directly identify and extract diffractions from common-shot gathers and to apply them for imaging. On the other hand, they have different geometrical characteristics in the dip-angle common-image gathers (DACIGs), which provides us an opportunity to extract diffractions. In this study, we present an efficient and accurate diffraction separation and imaging method in DACIGs using convolutional neutral network (CNN). The labeled data of DACIGs are generated using one pass of seismic modeling and migration for velocity models with and without artificial scatterers. Then, a simplified end-to-end CNN is trained to identify and extract reflections from the DACIGs with coupled reflections and diffractions. Next, two adaptive subtraction workflows are used to compute diffractions and stacked image. Numerical experiments for Marmousi-II and Sigsbee models demonstrate that the proposed method can produce accurate reflection and diffraction separation results in DACIGs and the stacked image show a good resolution for subsurface small-scale discontinuities. Note: This paper was accepted into the Technical Program but was not presented at IMAGE 2022 in Houston, Texas.