High-resolution stacking image based on common-image gathers

Seismic migration is an effective technique for reconstructing the subsurface geological structure. Focusing on common-image gathers (CIGs) produced by pre-stack depth migration, the imaging result is usually obtained by mean stacking CIGs from different shot-receiver pairs. For optimal results, the desired CIGs should exhibit consistent depth and waveform of the imaging wavelet across different angles or offsets. However, raw CIGs often suffer from misalignment due to inaccuracies in the migration velocity model and mad physical limitations. Additionally, factors such as complex topography, intricate subsurface structures, irregular acquisition geometries, and space-variant seismic wavelets contribute to uneven illumination and migration artifacts within CIGs. As a result, directly stacking CIGs does not yield high-resolution imaging. To address these challenges, we propose a novel imaging method based on CIGs. First, we analyze the high-resolution imaging process and identify residual depth, residual phase, and uneven illumination in CIGs as the primary factors affecting the stacking quality. To obtain high-resolution imaging, it is crucial to ensure that CIGs are flat and consistent in depth. Therefore, we have developed an improved dynamic programming method to flatten the CIGs. Next, to account for the uneven illumination across different angles and offsets, we propose an illumination-based weighted operator to optimize the CIGs, complemented by a similarity coefficient method for generating the stacking image. Finally, we present results from a synthetic model and field dataset to demonstrate the effectiveness of the proposed method. The results indicate that our approach significantly enhances imaging resolution and amplitude preservation, thereby improving imaging quality in complex exploration areas.