High-amplitude noise detection by the expectation-maximization algorithm with application to swell-noise attenuation

High-amplitude noise is a common problem in seismic data. Current filtering techniques that target this problem first detect the location of the noise and then remove it by damping or interpolation. Detection is done conventionally by comparing individual data amplitudes in a certain domain to a user-controlled local threshold. In practice, the threshold is optimally tuned by trial and error and is often changed to match the varying noise poweracrossthedataset.Wehavedevelopedanautomaticmethodtocomputetheappropriatethresholdforhigh-amplitudenoise detectionandattenuation.Themainideaistoexploitdifferences in statistical properties between noise and signal amplitudes to constructadetectioncriterion.Amodelthatconsistsofamixture of two statistical distributions, representing the signal and the noise,isfittedtothedata.Thenitisusedtoestimatetheprobabilityi.e.,likelihoodthateachsampleinthedataisnoisybymeans of an expectation-maximization EM algorithm. Only those samples with a likelihood greater than a specific threshold are consideredtobenoise.Theresultingprobabilitythresholdisbetter adapted to the data compared to a conventional amplitude threshold.Itofferstheuser,throughtheprobabilitythresholdvalue, the possibility to quantify the confidence in whether a large amplitude anomaly is considered as noise. The method is generic; however, our work develops and implements the method for swell-noise attenuation. Initial results are encouraging, showing slightly better performance than an optimized conventional methodbutwithmuchlessparametertestingandvariation.

[1]  Yang Liu,et al.  A 1D time-varying median filter for seismic random, spike-like noise elimination , 2009 .

[2]  Guillaume Cambois,et al.  Can We Surgically Remove Swell Noise , 1995 .

[3]  T. Elboth,et al.  De-noising Seismic Data In the Time-frequency Domain , 2008 .

[4]  George A. McMechan,et al.  Automatic editing of noisy seismic data , 1989 .

[5]  R. Redner,et al.  Mixture densities, maximum likelihood, and the EM algorithm , 1984 .

[6]  V. Hasselblad.  Finite mixtures of distributions from the exponential family , 1969 .

[7]  Robert Soubaras,et al.  Prestack Random And Impulsive Noise Attenuation By F-x Projection Filtering , 1995 .

[8]  Peter J. Rousseeuw,et al.  Robust Regression and Outlier Detection , 2005, Wiley Series in Probability and Statistics.

[9]  Alan Ryder,et al.  Swell-noise Attenuation Using an Iterative FX Prediction Filtering Approach , 2008 .

[10]  M. Baan,et al.  Local singular value decomposition for signal enhancement of seismic data , 2007 .

[11]  Mirko van der Baan,et al.  PP/PS Wavefield separation by independent component analysis , 2006 .

[12]  Athanasios Papoulis,et al.  Probability, Random Variables and Stochastic Processes , 1965 .

[13]  Tad J. Ulrych,et al.  Application of singular value decomposition to vertical seismic profiling , 1988 .

[14]  Steven Kay,et al.  Modern Spectral Estimation: Theory and Application , 1988 .

[15]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[16]  D. J. Verschuur,et al.  Adaptive surface-related multiple elimination , 1992 .

[17]  Peter J. Rousseeuw,et al.  Robust regression and outlier detection , 1987 .

[18]  Thomas Elboth,et al.  Flow and swell noise in marine seismic data , 2009 .