Getting accurate time-lapse information using geology-constrained simultaneous joint migrationinversion

We propose a simultaneous Joint Migration Inversion (SJMI) method for accessing time-lapse changes, which combines a joint time-lapse data processing strategy with the Joint Migration Inversion (JMI) method. Moreover, we extend it to include an L1-norm sparsity constraint on the reflectivities in a suitable transform domain and also naturally make a link with image structures and geological information by using a directional smoothing regularization on the velocities. We successfully demonstrate the effectiveness of the proposed method on a complex synthetic example based on the Marmousi model.