Abstract:Angle-Domain Common-Image Gathers (ADCIGs) is an important bridge connecting the pre-stack seismic data and reservoir characteristics, which is of great significance for seismic migration and reservoir characterization. Compared with Kirchhoff migration and one-way wave-equation migration, Reverse-Time Migration (RTM) is the most accurate imaging method for complex areas. Generating the RTM angle gathers with efficiency and robustness still is a challenge. In the paper, we will mainly discuss how to efficiently generate high -quality angle gathers with the optical flow method. The optical flow method can be applied for the wavefield direction estimation, during the process of RTM wavefield extrapolation. The Lucas-Kanade (LK) technique and the Horn-Schunck (HS) algorithm are two fundamental approaches. However, the LK method relies on local wavefield gradient and suffers from the singular phenomenon. The HS method is sensitive to noise, and the accuracy of the estimated wavefield direction is not high. We propose using the Combining Local and Global (CLG) optical flow method to estimate the wavefield direction. The accuracy of wavefield direction can be improved effectively by the proposed CLG optical flow method, especially for the reflected interface. And the proposed method is simple, efficient and convenient for parallel processing. Moreover, compared with the HS optical flow method, the proposed method has not much extra calculation. Then, we apply the wavefield decomposition method with the analytical wavefield and directional filtering to weaken the inaccurate estimation of optical flow method at the wave-front overlapping situation. In the analytical wavefield, the directional decomposition can be realized by just only the spatial Fourier transform of the wavefield. Then we calculate wavefield direction and extract angle gathers based on the decomposed wavefield. Furthermore, an effective normalization method and an improved least square division method are proposed to improve the accuracy and stability of angle estimation. Finally, theoretical and practical data illustrate the effectiveness of the proposed method.

RTM angle gathers based on the Combining Local and Global (CLG) optical flow method and wavefield decomposition method

Angle Domain Common Image Gather for Wave Equation

Angle-domain common-image gathers from plane-wave least-squares reverse time migration

Wave Equation Migration in Angle Domain

Z-99 MIGRATION AMPLITUDE CORRECTIONS IN ANGLE-DOMAIN USING BEAMLET DECOMPOSITION

Angle domain common image gathers from reverse time migration by combining the Poynting vector with directional decomposition

Extracting angle-domain common-image gathers in VTI media using ansiotropic-Helmholtz P/S wave-mode decomposition

An efficient least-squares reverse time migration in image domain

Implementation strategy of 3D reverse time migration on GPU/CPU clusters

Generating Azimuth–Reflection Angle Gathers from Reverse Time Migration Using the High-Dimensional Local Phase Space Approximation of Seismic Wavefields

Reflected wave least squares reverse time migration with angle illumination compensation

Reflection Angle-Domain Pseudoextended Least-Squares Reverse Time Migration Using Hybrid Regularization

Deprimary reverse time migration angle gathers with a stabilized Poynting vector

An efficient reverse time migration method using local nearly analytic discrete operator

Wavefield decomposition in arbitrary direction and an imaging condition based on stratigraphic dip

2-D and 3-D Image-Domain Least-Squares Reverse Time Migration Through Point Spread Functions and Excitation-Amplitude Imaging Condition.

Reverse Time Migration Based on Normalized Wavefield Decomposition Imaging Condition

Computing Angle Gathers From Imaging of Multiples Using a Poynting Vector Method for Improving Angular Illumination

Angle-dependent Image-Domain Least-Squares Migration Through Analytical Point Spread Functions

A computational method for wide-azimuth 3D dip-angle gathers using Gaussian beam migration

Elastic reverse time migration angle gathers using a stabilized Poynting vector without zero points within the wave propagation ranges