Bayesian wavefield separation by transform-domain sparsity promotion

Geophysics ◽  
2008 ◽  
Vol 73 (5) ◽  
pp. A33-A38 ◽  
Author(s):  
Deli Wang ◽  
Rayan Saab ◽  
Özgür Yilmaz ◽  
Felix J. Herrmann
Keyword(s):  
Bayesian Approach ◽  
Real Data ◽  
Transform Domain ◽  
Coherent Noise ◽  
Noise Sources ◽  
Wavefield Separation ◽  
Ground Roll ◽  
Successful Removal ◽  
Multiple Elimination ◽  
Made In

Successful removal of coherent-noise sources greatly determines seismic imaging quality. Major advances have been made in this direction, e.g., surface-related multiple elimination (SRME) and interferometric ground-roll removal. Still, moderate phase, timing, amplitude errors, and clutter in predicted signal components can be detrimental. Adopting a Bayesian approach, along with assuming approximate curvelet-domain independence of the to-be-separated signal components, we construct an iterative algorithm that takes predictions produced by, for example, SRME as input and separates these components in a robust manner. In addition, the proposed algorithm controls the energy mismatch between separated and predicted components. Such a control, lacking in earlier curvelet-domain formulations, improves results for primary-multiple separation on synthetic and real data.

Transform-domain noise synthesis and normal moveout-stack deconvolution approach to ground roll attenuation

Geophysics ◽  
2020 ◽  
Vol 86 (1) ◽  
pp. V15-V22
Author(s):  
Felix Oghenekohwo ◽  
Mauricio D. Sacchi
Keyword(s):  
Least Squares ◽  
Seismic Data ◽  
Field Data ◽  
Transform Domain ◽  
Coherent Noise ◽  
Least Squares Problem ◽  
Fourier Filtering ◽  
Ground Roll ◽  
Seismic Reflections ◽  
Strong Ground

Ground roll is coherent noise in land seismic data that contaminates seismic reflections. Therefore, it is essential to find efficient ways that remove this noise and still preserve reflections. To this end, we have developed a signal and noise separation framework that uses a hyperbolic moveout assumption on reflections, coupled with the synthesis of coherent ground roll. This framework yields a least-squares problem, which we solve using a sparsity-promoting program that gives coefficients capable of modeling the signal and noise. Subtraction of the predicted noise from the observed data produces data with amplitude-preserved reflections. We develop this technique on synthetic and field data contaminated by weak and strong ground roll noise. Compared to conventional Fourier filtering techniques, our method accurately removes the ground roll while preserving the amplitude of the signal.

scholarly journals SetRank: A Setwise Bayesian Approach for Collaborative Ranking from Implicit Feedback

2020 ◽  
Vol 34 (04) ◽  
pp. 6127-6136
Author(s):  
Chao Wang ◽  
Hengshu Zhu ◽  
Chen Zhu ◽  
Chuan Qin ◽  
Hui Xiong
Keyword(s):  
Bayesian Approach ◽  
Posterior Probability ◽  
State Of The Art ◽  
User Preferences ◽  
Implicit Feedback ◽  
Pairwise Preference ◽  
Entire List ◽  
Collaborative Ranking ◽  
Real World Datasets ◽  
Made In

The recent development of online recommender systems has a focus on collaborative ranking from implicit feedback, such as user clicks and purchases. Different from explicit ratings, which reflect graded user preferences, the implicit feedback only generates positive and unobserved labels. While considerable efforts have been made in this direction, the well-known pairwise and listwise approaches have still been limited by various challenges. Specifically, for the pairwise approaches, the assumption of independent pairwise preference is not always held in practice. Also, the listwise approaches cannot efficiently accommodate “ties” due to the precondition of the entire list permutation. To this end, in this paper, we propose a novel setwise Bayesian approach for collaborative ranking, namely SetRank, to inherently accommodate the characteristics of implicit feedback in recommender system. Specifically, SetRank aims at maximizing the posterior probability of novel setwise preference comparisons and can be implemented with matrix factorization and neural networks. Meanwhile, we also present the theoretical analysis of SetRank to show that the bound of excess risk can be proportional to √M/N, where M and N are the numbers of items and users, respectively. Finally, extensive experiments on four real-world datasets clearly validate the superiority of SetRank compared with various state-of-the-art baselines.

SIGNAL TRANSFER AND STOCHASTIC RESONANCE IN COUPLED NONLINEAR SYSTEMS

1999 ◽  
Vol 09 (06) ◽  
pp. 1159-1167 ◽  
Author(s):  
C. HAUPTMANN ◽  
F. KAISER ◽  
C. EICHWALD
Keyword(s):  
Stochastic Resonance ◽  
Harmonic Signal ◽  
Signal Transfer ◽  
Coherent Noise ◽  
Noise Sources ◽  
Bistable Systems ◽  
Coupled Nonlinear Oscillators ◽  
Initial Stage ◽  
Combined Action ◽  
Influence Of Noise

A model of coupled nonlinear oscillators is discussed, wherein Langevin-type bistable systems are combined with self-sustained oscillators. An external harmonic signal is coupled in a subthreshold manner into the bistable systems at the initial stage of the signal chain. Signal transfer through the oscillators is studied under the influence of noise. Different noise contributions, including spatially-incoherent and spatially-coherent noise sources are considered. Results reveal a stochastic resonance kind of behavior at different stages of the signal transfer, specifically the harmonic signal is transduced through the whole system of coupled oscillators. The combined action of spatially-incoherent and spatially-coherent noise exhibits constructive as well as destructive influences on signal amplification.

scholarly journals Bayesian Inference for the Difference of Two Proportion Parameters in Over-Reported Two-Sample Binomial Data Using the Doubly Sample

Stats ◽  
2019 ◽  
Vol 2 (1) ◽  
pp. 111-120 ◽  
Author(s):  
Dewi Rahardja
Keyword(s):  
Bayesian Inference ◽  
Closed Form ◽  
Bayesian Approach ◽  
Interval Estimation ◽  
Real Data ◽  
Simulation Studies ◽  
Posterior Distributions ◽  
Binomial Data ◽  
The Difference ◽  
Data Subject

We construct a point and interval estimation using a Bayesian approach for the difference of two population proportion parameters based on two independent samples of binomial data subject to one type of misclassification. Specifically, we derive an easy-to-implement closed-form algorithm for drawing from the posterior distributions. For illustration, we applied our algorithm to a real data example. Finally, we conduct simulation studies to demonstrate the efficiency of our algorithm for Bayesian inference.

VSP wavefield separation: Wave-by-wave optimization approach

Geophysics ◽  
2007 ◽  
Vol 72 (4) ◽  
pp. T47-T55 ◽  
Author(s):  
Emil Blias
Keyword(s):  
Eigenvalue Problems ◽  
Real Data ◽  
Data Sets ◽  
Optimization Approach ◽  
Arrival Times ◽  
Vertical Seismic Profiling ◽  
Wavefield Separation ◽  
Vsp Data ◽  
The Waves ◽  
Sampling Event

Waves propagating across a vertical seismic profiling (VSP) array may be distinguished by their differing arrival times and linear-moveout velocities. Current methods typically assume that the waves propagate uniformly with an unvarying wavelet shape and amplitude. These assumptions break down in the presence of irregular spatial sampling, event truncations, wavelet variations, and noise. I present a new method that allows each event to independently vary in its amplitude and arrival time as it propagates across the array. The method uses an iterative global nonlinear optimization scheme that consists of several least-squares and two eigenvalue problems at each step. Events are stripped from the data one at a time. As stronger events are predicted and removed, weaker events then become visible and can be modeled in turn. As each new event is approximately modeled, the fit for all previously removed events is then revisited and updated. Iterations continue until no remaining coherent events can be distinguished. As VSP data sets are typically not large, the expense of this method is not a significant limitation. I demonstrate with a real-data example that this iterative approach can lead to a significantly better VSP wavefield separation than that which has been available when using conventional techniques.

Physical Wavelet Frame Denoising

Geophysics ◽  
2003 ◽  
Vol 68 (1) ◽  
pp. 225-231 ◽  
Author(s):  
Rongfeng Zhang ◽  
Tadeusz J. Ulrych
Keyword(s):  
Seismic Data ◽  
Noise Suppression ◽  
Real Data ◽  
Wavelet Denoising ◽  
Seismic Signals ◽  
Wavelet Frame ◽  
New Approach ◽  
Design And Implementation ◽  
Ground Roll ◽  
Acoustic Processing

This paper deals with the design and implementation of a new wavelet frame for noise suppression based on the character of seismic data. In general, wavelet denoising methods widely used in image and acoustic processing use well‐known conventional wavelets which, although versatile, are often not optimal for seismic data. The new approach, physical wavelet frame denoising uses a wavelet frame that takes into account the characteristics of seismic data both in time and space. Synthetic and real data tests show that the approach is effective even for seismic signals contaminated by strong noise which may be random or coherent, such as ground roll or air waves.

scholarly journals Distributed Stochastic Search Algorithm for Multi-ship Encounter Situations

2017 ◽  
Vol 70 (4) ◽  
pp. 699-718 ◽  
Author(s):  
Donggyun Kim ◽  
Katsutoshi Hirayama ◽  
Tenda Okimoto
Keyword(s):  
Distributed Algorithms ◽  
Search Algorithm ◽  
Real Data ◽  
Stochastic Search ◽  
Automatic Identification ◽  
Identification System ◽  
Local Search Algorithm ◽  
Ship Collision ◽  
Stochastic Search Algorithm ◽  
Made In

Ship collision avoidance involves helping ships find routes that will best enable them to avoid a collision. When more than two ships encounter each other, the procedure becomes more complex since a slight change in course by one ship might affect the future decisions of the other ships. Two distributed algorithms have been developed in response to this problem: Distributed Local Search Algorithm (DLSA) and Distributed Tabu Search Algorithm (DTSA). Their common drawback is that it takes a relatively large number of messages for the ships to coordinate their actions. This could be fatal, especially in cases of emergency, where quick decisions should be made. In this paper, we introduce Distributed Stochastic Search Algorithm (DSSA), which allows each ship to change her intention in a stochastic manner immediately after receiving all of the intentions from the target ships. We also suggest a new cost function that considers both safety and efficiency in these distributed algorithms. We empirically show that DSSA requires many fewer messages for the benchmarks with four and 12 ships, and works properly for real data from the Automatic Identification System (AIS) in the Strait of Dover.

Noise removal by migration of time-shift images

Geophysics ◽  
2014 ◽  
Vol 79 (3) ◽  
pp. S105-S111 ◽  
Author(s):  
Sheng Xu ◽  
Feng Chen ◽  
Bing Tang ◽  
Gilles Lambare
Keyword(s):  
Time Lag ◽  
Real Data ◽  
Velocity Model ◽  
Noise Removal ◽  
Time Shift ◽  
Reverse Time ◽  
Coherent Noise ◽  
Data Set ◽  
Depth Migration ◽  
Time Migration

When using seismic data to image complex structures, the reverse time migration (RTM) algorithm generally provides the best results when the velocity model is accurate. With an inexact model, moveouts appear in common image gathers (CIGs), which are either in the surface offset domain or in subsurface angle domain; thus, the stacked image is not well focused. In extended image gathers, the strongest energy of a seismic event may occur at non-zero-lag in time-shift or offset-shift gathers. Based on the operation of RTM images produced by the time-shift imaging condition, the non-zero-lag time-shift images exhibit a spatial shift; we propose an approach to correct them by a second pass of migration similar to zero-offset depth migration; the proposed approach is based on the local poststack depth migration assumption. After the proposed second-pass migration, the time-shift CIGs appear to be flat and can be stacked. The stack enhances the energy of seismic events that are defocused at zero time lag due to the inaccuracy of the model, even though the new focused events stay at the previous positions, which might deviate from the true positions of seismic reflection. With the stack, our proposed approach is also able to attenuate the long-wavelength RTM artifacts. In the case of tilted transverse isotropic migration, we propose a scheme to defocus the coherent noise, such as migration artifacts from residual multiples, by applying the original migration velocity model along the symmetry axis but with different anisotropic parameters in the second pass of migration. We demonstrate that our approach is effective to attenuate the coherent noise at subsalt area with two synthetic data sets and one real data set from the Gulf of Mexico.

Surface‐related multiple elimination: Application on real data

1991 ◽  
Author(s):  
D. J. Verschuur ◽  
A. J. Berkhout ◽  
C. P. A. Wapenaar
Keyword(s):  
Real Data ◽  
Multiple Elimination
Sign in / Sign up

Export Citation Format

Share Document