Least-squares Reverse Time Migration for Blended Data with a Local Low-rank Constraint
K. Xiang, X. Chen, H. Chen and Y. Chen
Event name: 78th EAGE Conference and Exhibition 2016
Session: Seismic Imaging Theory - Advances in Least Squares Migration
Publication date: 31 May 2016
Info: Extended abstract, PDF ( 877.04Kb )
Price: € 20
The simultaneous-source shooting technique can accelerate field acquisition and improve spatial sampling but will cause strong interferences in the recorded data. Direct imaging of blended simultaneous-source data has been demonstrated to be an promising research field since there can be no need for the separation for blended sources before the subsequent traditional processing and imaging. The key issue in direct imaging of blended data is the strong artifacts in the migrated image. Although the least-squares migration can help reduce some artifacts, there are still residual artifacts in the image. Those artifacts mainly appears on the shallow part of the image and appears as spatially incoherent noise. We propose to apply the singular spectrum analysis (SSA) operator to attenuate the such artifacts during least-squares inversion. Considering that global SSA cannot deal with over-complicated data well, we propose to use local SSA in order to better remove noise and preserve steeply dipping components. The local SSA operator corresponds to a local low-rank constraint applied in the inversion process. The migration operator used in the study is the reverse time migration (RTM) operator. We use the Marmousi model example to show the superior performance of the proposed algorithm.