Three‐parameter Radon transform in layered transversely isotropic media

Abstract

A B S T R A C TRadon transform is a powerful tool with many applications in different stages of seismic data processing, because of its capability to focus seismic events in the transform domain. Three-parameter Radon transform can optimally focus and separate different seismic events, if its basis functions accurately match the events. In anisotropic media, the conventional hyperbolic or shifted hyperbolic basis functions lose their accuracy and cannot preserve data fidelity, especially at large offsets. To address this issue, we propose an accurate traveltime approximation for transversely isotropic media with vertical symmetry axis, and derive two versions of Radon basis functions, timevariant and time-invariant. A time-variant basis function can be used in time domain Radon transform algorithms while a time-invariant version can be used in, generally more efficient, frequency domain algorithms. Comparing the time-variant and time-invariant Radon transform by the proposed basis functions, the time-invariant version can better focus different seismic events; it is also more accurate, especially in presence of vertical heterogeneity. However, the proposed time-invariant basis functions are suitable for a specific type of layered anisotropic media, known as factorized media. We test the proposed methods and illustrate successful applications of them for trace interpolation and coherent noise attenuation.

A High-Precision Method for Extracting Polarization Angle Under the Condition of Subsurface Wavefield Aliasing

“…where, the long window length M = 5N and the short window length N = k/ i (t), k is the weight, and i (t) is the instantaneous dominant frequency of the i-th moment after the generalized S transform of the three-axis vibration signal [10], which refers to the frequency point with the highest energy among the wide spectrum corresponding to each moment after the generalized S transform. With the optimal matching impedance, in the near field of the blasting point, the moment the first break arrives the instantaneous dominant frequencies of the three axes are consistent [11].…”

Approximations for traveltime, slope, curvature, and geometric spreading of elastic waves in layered transversely isotropic media

“…Figure 8d shows the results. Figure 8e and 8f show the separated primary P and PS reflections, using an anisotropic Radon transform (Abedi et al, 2019c). To summarize, Figure 8 shows a CMP processed for P and PS primary reflection data, with potential application for stacking and AVO analysis of each wave mode.…”

A Multidirectional Deep Neural Network for Self-Supervised Reconstruction of Seismic Data

“…Reconstruction of the missing data parts from the existing data is based on assuming a form of continuity for specific seismic events, using simplifying assumptions about the events' shape, or finding such continuity from the existing data itself. Classical data reconstruction methods use prediction filters in the frequency domain (e.g., [1]- [3]), curvelet transform (e.g., [4], [5]), seislet transform (e.g., [6], [7]), Radon transform (e.g., [8]- [11]), and inversion (e.g., [12], [13]). Recent data reconstruction methods that use data-driven approaches include dictionary learning (e.g., [14], [15]), support vector regression (e.g., [16], [17]), and deep learning (e.g., [18]- [24]).…”