Abstract
spectrum, which is pretrained using synthetic traces and STF labels. Second, based on the pretrained STFNet, we design a Time-frequency (TF) transform is a commonly used tool for DASTFA model for transferring to field data. Note that the syn-geologic structure interpretation and attenuation estimation, thetic traces are generated using well-log data and horizon at the mainly including the linear and nonlinear methods. The former study area, located in the Ordos Basin, China, which can reduce consists of short-time Fourier transform, continuous wavelet the gap between synthetic and field data. Afterward, we use syn-transform, S-transform, etc., which can be efficiently implethetic traces with their STF labels for model pretraining in the mented. However, they suffer from the Gabor uncertainty prinfirst step and field data without STF labels for model transferciple, which limits their TF readability. The latter can effectively ring in the second step. Here, model transferring indicates trans-enhance TF readability; nevertheless, it has several unavoidable ferring the pretrained model parameters based on field traces. drawbacks, such as low computational efficiency and parameter Finally, to test the validity and effectiveness of DASTFA, we selection. For seismic sparse TF (STF) analysis (STFA), we deapply it to synthetic and field data. Moreover, we test the availvelop a domain adaptation-based STFA (DASTFA) model. This ability of seismic attenuation estimation by adopting the sugmodel mainly contains two steps. First, we suggest an STF net-gested model for 3D poststack field data volume, which work (STFNet) for mapping a 1D seismic signal to a 2D STF achieved encouraging results.
| Original language | English |
|---|---|
| Pages (from-to) | B187-B198 |
| Journal | Geophysics |
| Volume | 89 |
| Issue number | 3 |
| DOIs | |
| State | Published - 1 May 2024 |
Fingerprint
Dive into the research topics of 'Domain adaptation-based sparse time-frequency analysis and its application on seismic attenuation estimation'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver