... for estimating and correcting water velocity changes on 4D seismic datasets. Each sail line is migrated independently and Common Depth Point (CDP) gathers are output for the overburden along a subsurface strip for both vintages. The 4D approach consists of performing cross-correlations using collocated CDPs...

... probability hyperparameter lellouch dataset Automatic microseismic event detection in downhole DAS data through convolutional neural networks: A comparison of events during and post-stimulation of the well Paige Given*, Fantine Huot, Ariel Lellouch, Bin Luo, Robert G. Clapp, Biondo L. Biondi, Stanford...

... management. However, it remains unclear how effectively remotely sensed datasets can be integrated to close the water balance and estimate changes in groundwater storage at different spatial scales. Because many remote sensing platforms provide free near real-time and data at high spatial (<1km – 10km...

... deep learning international application neural network reservoir characterization prediction figure 6 figure 5 applied geoscience american association society expanded abstract dataset figure 8 geophysics waveform inversion fwi exploration geophysicist inversion bandwidth extension...

... datasets required by such algorithms. We present a three-fold solution that greatly alleviates the memory footprint and computational cost of 3D MDC by leveraging a combination of i) distance-aware matrix reordering, ii) Tile Low-Rank (TLR) matrix compression, and iii) computations in mixed floating-point...

... starting velocity model. However, low-frequency components of a 3D land dataset used in this study are dominated by noises and a legacy velocity model in the survey area is not existed. To overcome these problems, a two-stage FWI workflow has been implemented to obtain a credible shallow velocity model...

... in effectively finding, characterizing, and maturing multiple and different types of targets in seismic volumes. We used the Teapot Dome 3D seismic survey as input for demonstrating the use of the Model-Grid and unique approach of the Relative Geological Time Model on a US onshore dataset. From these models, we...

... association annual logging symposium dataset machine learning high resolution fast inversion borehole boundary artificial intelligence ml-based deconvolution method exploration geophysicist applied geoscience ML-based Deconvolution Method for High Resolution Fast Inversion of Induction Log Data...

... containment information integration american association dataset containment assessment seismic inversion society software engineering climate change integrated workflow applied geoscience energy 10 Integrated workflow for characterization of CO2 subsurface storage sites Noémie Pernin, Cyrille...

... dataset, as these data were shot when a platform was in place over the crest of the field. NATS data were shot previously with no obstruction, however the two surveys have different azimuthal and angular coverage. To understand the impact and illumination of each dataset, we undertook a modeling study...

... for the exploration of low-relief structures. The inversion of the large TEM dataset is carried out with a hybrid inversion approach combining deep learning and physics-driven least-squares inversion. The procedure provides a sharp mapping of the shallow BoS resulting in the improvement of the seismic images...

... We combine the ideas of transfer learning and semantic segmentation of seismic facies in a dataset-to-dataset transfer learning research problem with the goal of reducing the number of inputted labeled seismic slices while also improving computational cost and accuracy of predicting geological...

... learning exploration geophysicist diffracted wavefield reflection transfer applied geoscience international gajewski prediction energy 10 dataset american association convolutional autoencoder university Transfer learning seismic and GPR diffraction separation with a convolutional neural...

... Every seismic dataset has its particular characteristics guided mainly by the properties of the subsurface, the data acquisition parameters (the survey), and the often unique noise conditions it experiences. Capturing such characteristics in a neural network model for the efficient application...

... This study uses ImageNet pretrained convolutional neural networks (CNNs), VGG11 and ResNet18 models to predict carbonate rock and pore types on a small dataset of 66 thin sections. We subsequently overlay Gradient Weighted Class Activation Maps (Grad-CAM) on top of the original thin section...

... azimuth (WAZ) streamer data is available. We present the results of our model-building workflow and learnings in this paper centered around FWI using a single WAZ survey at Campeche Block 9. international dataset applied geoscience inversion fwi image reservoir characterization boundary...

... dataset birnie procedure alkhalifah applied geoscience noise suppression pixel international exploration geophysicist application upstream oil & gas artificial intelligence dataset epoch society american association energy 10 international conference Boosting self-supervised blind...

... of a petrographic dataset are presented. The StyleGAN2 architecture was selected to train a set of 10070 petrographic images which were divided into four categories: plutonic, volcanic, sedimentary, and metamorphic rocks. This model achieved a state-of-the-art FID (Fréchet Inception Distance) score of 12.49...

... dataset. The trained network is able to instantaneously estimate the subsurface property change for any new monitoring dataset without conventional velocity model building and imaging. A deep learning architecture with a new multi-branch design with different filtering sizes is implemented for better...

... geology and drilling targets. In this paper, we apply machine learning technology in predicting faults and horizons in a structurally and geologically complex onshore Texas dataset. By employing ML technology through convolutional neural networks (CNNs) trained on real data we predict multiple layers...