SEISMIC DATA INTERPOLATION WITH CURVELET DOMAIN SPARSE CONSTRAINED INVERSION

To improve the accuracy of seismic data processing, the irregular, missing seismic data need interpolation. Retrieving the missing data can decrease the error in the prediction of multiple and the aliasing of image etc. The iterative threshold method based on l(1) norm optimization inversion with sparsity constraint can achieve better result with increase of the sparsity of the inversion parameters. Take the advantage of the sparse representation of seismic data in the Curvelet domain, we can get a better result when performing a l(1) norm optimization problem in seismic data analysis. In this paper, we apply this method in interpolation of missing seismic data, and using the Curvelet threshold iterative method after NMO correction, which make the data sparser. By comparing the results of the interpolation with and without NMO correction, we confirm that the Curvelet threshold iterative method can get a better effect in seismic missing data interpolation. Its results are more close to the initial model as well after the process of NMO correction.