Bandwidth extension using 3D structure preserving sparse-blind deconvolution
Stephen K. Chiu, Duggan, Yifeng Jiang · 2018
We propose a new 3D structure-preserving sparse blind deconvolution in extending data bandwidth. This method iteratively estimates a source wavelet and solves reflectivity series using a sparse deconvolution constrained by 2D/3D structure preserving filter. A multidimensional Cadzow filter is selected for its robustness to attenuate the noise and to better condition the deconvolution process. The extension of the sparse blind deconvolution to 3D further improves the wavelet estimation, yields a better and more robust sparse deconvolution, and avoids the pitfalls of conventional sparse blind deconvolution in creating spatially discontinuous seismic events. In addition, the data after this proposed method maintains the same phases as the input data. The effectiveness of the proposed method will be validated with synthetic data and field data from Permian Basin. Presentation Date: Monday, October 15, 2018 Start Time: 1:50:00 PM Location: Poster Station 19 Presentation Type: Poster