摘要

The high-resolution information carried by small-scale geological discontinuities in subsurfaces such as faults, fractures, watercourses, and coarse rock boundaries can be fully exploited.In this study, a separation and imaging method of diffracted waves in dip-angle common-image gathers (CIGs) was developed, based on a hybrid-domain high-resolution Radon transform, which is solved by a sparse constrained preconditioned conjugate gradient algorithm.This method uses the morphological differences between reflected and diffracted waves, that is the events of reflection are concave curves with stable phase apex while the diffraction are quasi-linear events, to separate and image diffractions in dip-angle CIGs.First, the preconditioned operator was introduced into the traditional Radon transform in the hybrid domain; then, a new time-varying sparse weight of the Radon model was constructed.Finally, separation and imaging of the diffractions in the dip domain CIGs were conducted.Tests on synthetic and field data demonstrated that, compared with the least-square Radon transform and with the high-resolution Radon transform method in the frequency domain, the proposed method removed the reflections more effectively.The proposed method can improve the imaging quality of small-scale geological bodies such as faults.