..., we introduce a novel perceptual quality-based model training framework to objectively generate multiple labels for model training to enhance reliability, while avoiding massive annotations. Specifically, we first select a subset of samples with low perceptual quality scores ranked by statistical...

... area into subsets and connecting them recursively until the entire area is completed. The connecting equations are derived using the Dyson equation, widely used in quantum mechanics. The wavefield calculated at the target and receivers is numerically equivalent to obtained with wave propagation...

... volumes at a time, primarily selected based on the interpreter preference. Fortunately, machine learning technology provides the opportunity to quantitatively use many frequency volumes. We analyze twelve spectral magnitude components using multivariate feature selection techniques. The chosen subsets...

... practical studies, it is unfeasible to run fluid flow simulations and predict the reservoir production for hundredths of reservoir realizations due to the computational cost of dynamic fluid flow simulations. For this reason, we propose a statistical method for the selection of a subset of reservoir models...

... moveout (LMO) correction, the original seismic record can be divided into multiple subsets with a reasonable set of the time window and sliding length, and each subset contains only 1~2 complete surface-wave events. The phase shift method is used to obtain the dispersion image of each subset. Through...

... ABSTRACT Selecting appropriate attributes is a crucial part of the seismic facies classification workflow for reducing computational cost and reasonable model building. For supervised learning, the best attribute subset can be built by selecting input attributes that identify...

... be explored to optimize and coordinate interpretation efforts. In this study, a field 3D dataset was geometrically decomposed into near maximum-coupled (so-called radial) and near null-coupled (so-called tangential) subsets and inverted using a 3D approach. The results indicate that these two models may...

.... With the application of triangular filter, the time-frequency cube is then divided into frequency-amplitude subsets of different frequency ranges in accordance with sequence thickness of different orders derived from well logging interpreted thickness distribution and related dominant frequency of sequence orders...

... Type: ORAL reservoir characterization seg seg international exposition annual meeting qc methodology dc resistivity data electrode normalized misfit subset dataset resistivity artificial intelligence inversion geophysics statistical data duality control methodology misfit machine...

... in ‘smart’ subsets that emerge naturally from the acquisition type, and using focal operators designed for those subsets. In this way, each subset can be deblended independently from the rest. We define such subsets for the case of a marine towed multi-streamer acquisition and test the deblending...

... of the objective function, gradient-based inversion methods are used combined with multiscale strategies to improve convergence to global minimum. We discuss the formulation and application of three data-domain multi-scale approaches that address the non-linearity of the objective function by taking a subset...

... with the known spatial heterogeneity in shale gas reservoirs. variation dipole dispersion walsh placement poisson shale gas reservoir characterization producer well vertical pilot well reference state complex reservoir upstream oil & gas subset ticonstant vertical depth fracture...

... with the matrix factorization, makes them still inappropriate for solving 3D elastic problems. approximation linear system multiple-right hand side strategy requirement cgmn method subset operation seed system mrh strategy seed method equation upstream oil & gas solver gordon...

... the computation cost. Considering the often significant numbers of sources (and receivers) in 3D seismic data acquisition, we propose a method to largely reduce the simulation time per iteration by using a reasonably small subset of sources instead of the full volume. We choose the subset by randomly picking...

... Summary A known mineralized zone was successfully mapped in high resolution with the Orion 3D omnidirectional 3D DCIP system. Subsets of the original dataset are used to examine the resolution and mapping capabilities of other approaches to 3D DCIP acquisition. The results show that the high...

... geometries and found that irregular node configurations worked better than regular lines when the node positions were not repeated. Implications of these findings on future time-lapse survey designs are discussed. upstream oil & gas receiver decimation subset nrm value dataset production inc...

... ABSTRACT We formulate 3 alternative approaches to the Laplace-Fourier domain formulation of waveform inversion: i, ) A conventional L, 2, misfit function, and ii, ) and iii, ) phase-only misfits derived from the logarithmic wavefield residuals. In ii, ) we use multiple subsets of frequencies...

... that provided by simulating all the shots in a survey simultaneously, which requires a seldom-used acquisition geometry. speedup essfwi computational cost velocity model subset inversion iteration rbfwi geometry fwi reservoir characterization msg exxonmobil upstream research company upstream...

... ABSTRACT RTM Migrations of different subsets of the SEG Advanced Modeling (SEAM) project data show the following features: • Sparse shots with full receiver coverage do a surprisingly good job of imaging subsalt; • Noise from multiples is significant only in a very few locations in the Classic...

... vector regression machine subset singular value decomposition well logging well log data core measurement Predicting permeability from well log data and core measurements using Support Vector Machines Siamak Nazari*, UC Berkeley, Heidi A. Kuzma, East Donner Research LLC., James W. Rector III, UC...