The improved mathematical morphological filtering for low-frequency noise attenuation

Source abstract

Low-frequency noise is ubiquitous in seismic data, so the attenuation of this noise is important to improve the S/N. Traditional approaches, such as high-pass filtering and empirical mode decomposition (EMD), use the differences in frequency to distinguish between useful signal and noise. However, in some cases, the traditional methods are limited or even invalid when the differences are too small to be separated or even the signal and noise share the same frequency, which is a troublesome problem in building the velocity model. Unlike frequency band filtering, the traditional MMF method uses the differences in morphological scale to separate signal and noise. However, the gradient of SE window is unadjustable. For this reason, we developed an improved method contained flexible window of Gauss function, which can preserve more low-frequency signal and suppress more noise than the original MMF approach, EMD approach and high-pass filtering in synthetic and field seismic examples. Presentation Date: Tuesday, October 16, 2018 Start Time: 9:20:00 AM Location: Poster Station 9 Presentation Type: Poster