Deep Unsupervised 4-D Seismic 3-D Time-Shift Estimation With Convolutional Neural Networks

Dramsch, Jesper Sören; Christensen, Anders Nymark; MacBeth, Colin; Lüthje, Mikael

doi:10.1109/tgrs.2021.3081516

Cited by 8 publications

(2 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Не вызывает сомнения польза применения искусственного интеллекта, в том числе нейросетевых алгоритмов, для решения задач математической картографии и геоинформационного моделирования [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29].…”

Section: Introductionunclassified

Development and application of neural network technology in solving geodynamic problems

Maximova

2023

jour

View full text Add to dashboard Cite

The present study is aimed at searching the most effective, easy-to-use method for seismic zoning, in order to consider the ongoing environmental change in real time. The study demonstrates the relevance and necessity of tackling this problem, provides characteristics and peculiarities of current microzonation methods, and discusses the problem of seismic microzoning. The paper highlights traditional and neural network approaches as two directions in solving the problem, and determines their advantages and disadvantages. In addition, the paper provides significant arguments for applying the neural network approach and defines the perspectives for the solution. An intermediate problem of geophysics is suggested to be solved using the neural network approach. The current study results involved searching networks able to provide sufficient accuracy for obtaining pictures of geological sections and conducting analysis of predicted results for a number of neural networks. Multilayer perceptron is considered to give the most reliable results. The future work is supposed to develop an algorithm for building a map of seismic microzonation by means of neural network technology.

show abstract

Section: Introductionunclassified

Development and application of neural network technology in solving geodynamic problems

Maximova

2023

jour

View full text Add to dashboard Cite

show abstract

“…These methods, however, often necessitate extensive labeled datasets for training, a significant impediment given the rarity and unpredictability of seismic anomalies. The current zenith of anomaly detection methodology in seismology is epitomized by unsupervised learning algorithms, which do not require labeled data and are adept at handling the data-rich but label-sparse reality of seismic monitoring (Dramsch, Christensen, MacBeth, & Lüthje, 2021). The Isolation Forest algorithm has emerged as a leading-edge tool, designed to isolate anomalies rather than model the 'normal' data, an approach that is inherently suitable for the seismic domain where 'normal' is a fluid and elusive concept (Heigl et al, 2021;Lesouple, Baudoin, Spigai, & Tourneret, 2021).…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Machine Learning for Seismic Anomaly Detection: Isolation Forest Algorithm Application to Indonesian Earthquake Data

Airlangga

2023

SCI TECH ED MATH

View full text Add to dashboard Cite

Indonesia's position on the seismically active Pacific "Ring of Fire" necessitates advanced methods for earthquake detection and preparedness. This paper introduces an application of the Isolation Forest algorithm—an unsupervised machine learning technique—for detecting seismic anomalies in Indonesia's complex geotectonic landscape. Unlike traditional methods that rely on predefined thresholds or patterns, the Isolation Forest algorithm isolates anomalies based on their rarity and distinctness without the need for labeled data. We applied this algorithm to a comprehensive dataset from the Indonesian Meteorology, Climatology, and Geophysical (BMKG) Agency, featuring a decade's worth of seismic events, to identify outliers that may signify potential seismic hazards. Our findings reveal that the Isolation Forest algorithm effectively identifies seismic anomalies, with 874 out of tens of thousands of events flagged as statistically significant outliers. A comparative analysis with traditional seismic anomaly detection models highlights the robustness of the Isolation Forest algorithm in handling the high-dimensional and noisy nature of seismic data, emphasizing its superiority in detecting subtle anomalies

show abstract

Machine learning-based time-lapse 1D seismic full-waveform inversion with efficient training data generation in a carbon capture and storage monitoring

Kim,

Park,

Seol

et al. 2024

Geoenergy Science and Engineering

View full text Add to dashboard Cite

Deep Unsupervised 4-D Seismic 3-D Time-Shift Estimation With Convolutional Neural Networks

Cited by 8 publications

References 44 publications

Development and application of neural network technology in solving geodynamic problems

Development and application of neural network technology in solving geodynamic problems

Unsupervised Machine Learning for Seismic Anomaly Detection: Isolation Forest Algorithm Application to Indonesian Earthquake Data

Machine learning-based time-lapse 1D seismic full-waveform inversion with efficient training data generation in a carbon capture and storage monitoring

Contact Info

Product

Resources

About