trustme.bro/r/…
✓ checked
trust me, bro:
here is the receipt.
the claim
Machine learning can be applied to raw seismic data and CMP stacking for seismic interpretation.
the verdict
SUPPORTED
the evidence backs this
refutedsupported
the weight of evidence
3 sources for · 0 against

Peer-reviewed literature demonstrates that machine learning can be applied to raw seismic data and common midpoint (CMP) stacking for seismic interpretation.

Evidence for · 3
2021 · cited by 0
Abstract Amplitude variation with offset (AVO) analysis has become a crucial step in determining gas well locations. However, large amounts of time and effort are required to confirm AVO anomalies, and interpretations can be inconsistent. To avoid the need for the long processes involved in conventional manual analysis, we developed an automatic AVO analysis method for common midpoint (CMP) gathers through using machine learning (ML) with a convolutional neural network (CNN). To deal with complicated seismic data, the network was constructed based on VGG16 network architecture, which includes 16 layers. The resulting CNN-based algorithm was applied to two sets of three-dimensional (3D) seismic data acquired off the east coast of South Korea. One dataset, which was obtained over gas reservoirs and confirmed to show multiple AVO class III anomalies by comparison with geophysical well logging data, was used for training and evaluation of the proposed CNN-based algorithm. The resulting trained model was tested using the second dataset, which was obtained over an area near the gas reservoirs with a different depositional environment. To demonstrate the applicability of the model to raw and final migrated CMP gathers, AVO class III anomalies predicted by the ML-derived analysis were confirmed by manual AVO analysis of the test data.
See more details
The analysis

rails:sufficiency:supported:single_source:for=1+2p:against=0+0p | v55:sufficiency

More for · 2
cited by 0
Applications de l’apprentissage profond pour le traitement et l’interprétation sismique Acquérir des connaissances sur la géologie de la subsurface terrestre grâce à l’imagerie sismique est un processus long et parfois fastidieux. De nombreux algorithmes sont utilisés pour transformer le signal, atténuer le bruit et aider à interpréter l’image. Ces algorithmes sont conçus par des experts et nécessitent d’être soigneusement paramétrés. De plus, de nombreuses tâches doivent être effectuées manuellement par les géoscientifiques lorsque les algorithmes ne parviennent pas à fournir de bons résultats. Ces dernières années, l’apprentissage profond, un sous-domaine de l’intelligence artificielle, a pris une grande importance. Il a été montré que les modèles d’apprentissage surpassent les algorithmes traditionnels dans de nombreuses applications à travers un grand nombre de disciplines scientifiques. Ils permettent également d’automatiser certains processus qui n’étaient jusque-là réalisables que par des humains. Cependant, il peut être difficile de remplir les conditions nécessaires pour exploiter leur potentiel. Dans cette thèse, nous identifions les principaux obstacles à l’utilisation de l’apprentissage profond, notamment ceux de l’incertitude sur l’interprétation des données et de la dépendance de l’apprentissage en exemples fournis par des experts, et proposons une série de méthodologies visant à les surmonter. Nous démontrons la validité et la faisabilité de nos méthodes sur un ensemble de problèmes d’interprétation et de traitement sismique.
cited by 0
c volume in a fixed coordinate ( IL , XL , or Z ) is called a slice ( Figure 1 b). Figure 1. Open in a new tab ( a ) Seismic Volume, ( b ) Slice (cut on the seismic volume) on a fixed IL coordinate. Given that a seismic volume can be considered a sequence of subsurface images, we propose two different approaches for seismic data clustering: Pointwise Data Clustering and Spatial Groups Data Clustering. The two alternatives are represented in Figure 2 . Figure 2. Open in a new tab Proposed seismic data clustering approaches. Pointwise Data Clustering analyzes the seismic volume using each data point o I L ,   X L ,   Z as a sample. These samples can be described with multiple seismic attributes commonly extracted from the amplitude seismic data, such as energy, frequency, phase, among others. With the Seismic Volume usually requiring Gbs of storage, the addition of multiple seismic attributes for an accurate interpretation can escalate the volume of data to a prohibitive size for most Machine Learning algorithms. Therefore, we propose feature selection [ 33 ] to eliminate irrelevant or redundant seismic attributes, resulting in a significant reduction in the data size. Specifically, 28 seismic attributes are reduced to a subset of only 12 attributes employing the rankings provided by four feature selection algorithms: Principal Component Analysis [ 34 ], Principal Feature Analysis [ 35 ], Variance Threshold [ 36 ], and Feature weighting k-Means [ 37 ]. After selecting the best subset of seismic attributes, classical clustering algorithms are applied to the data, obtaining post-processed groups to enhance their homogeneity inside the seismic volume. These groups represent points in this multivariate seismic volume with similar seismic attributes values. Spatial Groups Data Clustering, in contrast, defines a processing pipeline that employs image segmentation algorithms to define preliminary spatial groups of seismic data denominated segments. This approach begins by di
Everything we examined (3)
This check searched the claim as stated. It did not run a separate search for evidence against it.
  1. Machine learning derived AVO analysis on marine 3D seismic data over gas reservoirs near South Koreapeer-reviewedno side taken
  2. Deep learning for seismic data processing and interpretationpeer-reviewedno side taken
  3. Unsupervised Machine Learning Applied to Seismic Interpretation: Towards an Unsupervised Automated Interpretation Tool - PMCofficial-recordno side taken
This receipt carries no identity, shared or not. Sharing publishes your connection to it, not your data.
Check your own claim
Challenge the receipt
trust me, bro: win the argument, pass the class, survive peer review.
This receipt is an automated verdict against our published method · not an opinion about any author or publication.
Terms · Privacy · How verdicts work · Dispute this receipt