Adaptive prediction filtering in - - domain for random noise attenuation using regularized nonstationary autoregression |
Equation 5 shows that one sample in - domain can be predicted by the samples in adjacent traces with weight coefficients , which is time- and space-varying. The equation assumes that the seismic data only consist of plane waves and random noise that corresponds to a least-squares error.
Figure 1a shows a 2D space-causal APF structure, which is time-noncausal filter. White grids stand for prediction samples and the dark-grey grid is the output (or target) position, while light-grey grids are unused samples. The filter size of the space-causal APF is . Meanwhile, space-noncausal APF (Figure 1b) has a symmetric structure along time and space axes. The filter size of the space-noncausal APF is . The 3D - - APF also has space-causal or space-noncausal structure, Figure 2 shows the noncausal one. In a 3D seismic datacube, the plane events can be predicted along two different spatial directions. A 2D - APF will have difficulty preserving accurate plane waves because it only uses the information in or direction, however, a 3D - - APF provides a more natural structure. - - adaptive prediction filtering for random noise attenuation follows two steps:
1. Estimating 3D space-noncausal APF coefficients
by solving the regularized least-squares
problem (equation 4 or 5 in 2D):
2pt
2. Calculating noise-free signal
according to
2pt
causal2d,noncausal2d
Figure 1. Schematic illustration of a 2D - APF. A space-causal filter (a) and a space-noncausal filter (b). |
---|
noncausal3d
Figure 2. Schematic illustration of a 3D - - space-noncausal APF. |
---|
Adaptive prediction filtering in - - domain for random noise attenuation using regularized nonstationary autoregression |