![[pdf]](icons/pdf.png){kind=icon}
|
|
|
|
| Random noise attenuation by a selective hybrid approach using f-x empirical mode decomposition | |
|
EMD. The ability of
EMD to preserve horizontal events is very strong, however, it is very sensitive to dipping events. Even after removing many IMFs in the
domain, the horizontal events can still be preserved. However, even removing one or two IMFs in the
domain, the dipping events can be totally removed. In order to solve the problem of
EMD in dealing with complex structure that contains dipping events and at the same to improve the accuracy for a selected dipping-events retriever, we have proposed a novel and general hybrid denoising framework, which fully utilizes the horizontal-preservation capability of
EMD in dealing with non-stationary seismic data and the dipping-preservation capability of the selected dipping-events retriever. As a tutorial,
SSA is selected to be combined with
EMD in this paper.
A selective hybrid strategy is also proposed to maximize the effectiveness of
EMD and the processing efficiency. The current selective hybrid approach is based on manually selected processing windows, which is inconvenient for implementation when seismic profile is over complicated. An automatic way to detect the specific windows containing complex structure and to implement the selective hybrid denoising approach is the topic of current research. From the synthetic and field data examples, it's obvious that, the proposed selective hybrid approach can effectively obtain better results compared with both
EMD and the selected dipping-events retriever.
|
|
|
|
| Random noise attenuation by a selective hybrid approach using f-x empirical mode decomposition | |
|