Article
Geochemistry & Geophysics
Chao Song, Tariq A. Alkhalifah
Summary: Wavefield Reconstruction Inversion (WRI) is a method that tackles the cycle skipping problem in Full-Waveform Inversion (FWI) by formulating an optimization problem that satisfies the Helmholtz equation while fitting the recorded data. Compared to finite-difference methods, the PINN-based WRI method does not require matrix inversions and offers greater flexibility and adaptability.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Multidisciplinary
Yong Li, Gang Cheng, Chang Liu
Summary: In this paper, a bearing fault diagnosis algorithm based on spectrum characteristics and MOMEDA is proposed to address the difficulty of diagnosing bearing faults under strong noise interference. By splitting the original time-domain signal and adopting spectrum fusion, the algorithm effectively extracts the characteristic frequency of faulty bearings under stronger noise interference.
Article
Geochemistry & Geophysics
Yuzhao Lin, Huaishan Liu, Jian Sun, Lei Xing
Summary: Wavefield Reconstruction Inversion (WRI) is a novel seismic inversion method that reduces nonlinearity and improves robustness compared to conventional techniques. By incorporating the Unbalanced Optimal Transmission (UOT) distance measurement method, this study addresses inaccuracies in time-domain WRI and ensures accurate wavefield reconstruction. The method iteratively solves the inversion problem, accurately measures the misfit, and extends the source for wavefield reconstruction based on an initial defective model, without excessive additional computation.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Qingqing Li, Li-Yun Fu, Ru-Shan Wu, Qizhen Du
Summary: Q-compensated reverse-time migration (Q-RTM) is efficient for seismic imaging, but storing the source wavefield on disk has been a bottleneck. A reversible hybrid absorbing boundary condition is proposed in this letter to address the storage issue and dramatically improve computational efficiency.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Engineering, Electrical & Electronic
Dingcheng Ji, Jing Lin, Fei Gao
Summary: Conventional full wavefield scanning techniques for acquiring Lamb waves in CFRPs suffer from high time and energy consumption. To address this issue, we proposed a novel autoencoder-based method that can accurately reconstruct high-sparsity wavefield data.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Mechanical
Jun Young Jeon, Yinan Miao, Gyuhae Park, Eric Flynn
Summary: This paper presents a compressive sensing-based high speed full-field laser scanning technique for structural damage detection. By utilizing compressive sensing, the efficiency of the damage detection and visualization process can be improved with high accuracy.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Riwaj Byanju, Stefan Klein, Alexandra Cristobal-Huerta, Juan A. Hernandez-Tamames, Dirk H. J. Poot
Summary: This study proposes a theoretical metric called TEUSQA to evaluate the time efficiency of accelerated Quantitative MRI sequences. The time efficiency predicted by TEUSQA is validated through simulations and experiments. The study also suggests that low-discrepancy sampling can generate good undersampling patterns.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Geochemistry & Geophysics
Yuzhao Lin, Huaishan Liu, Lei Xing, Haoran Lin
Summary: Full waveform inversion (FWI) evaluates subsurface media properties by minimizing the misfit between synthetic and observed data. This study introduces a new form of FWI, called Wavefield Reconstruction Inversion (WRI), which combines theoretical and measurement errors to handle nonphysical data, addressing some practical issues.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Zhan Xu, Keith A. A. Michel, Christopher M. M. Walker, Collin J. J. Harlan, Gary V. V. Martinez, Jeremy W. W. Gordon, Hsin-Yu Chen, Daniel B. B. Vigneron, James A. A. Bankson
Summary: MoReFUn technique is introduced as an effective and reliable imaging acceleration method for metabolic imaging using hyperpolarized [1-C-13]-pyruvate, which resolves spatial aliasing by introducing constraints based on pharmacokinetic model. The technique showed successful reconstruction in numerical digital phantom, in vitro dynamic multispectral phantom, and prostate cancer patient data sets, with high accuracy and minimal increase in error.
MAGNETIC RESONANCE IN MEDICINE
(2023)
Article
Engineering, Aerospace
Min-Seok Kang, Jae-Min Baek
Summary: This article presents an incremental SAR imaging approach based on the estimation of sensing dictionary matrix to address the SAR azimuth ambiguity. The proposed method achieves significantly better SAA reduction compared to conventional compressive sensing (CS)-based approaches in terms of image quality and computational efficiency. The method establishes a signal parameterization scheme and utilizes an incremental imaging approach to overcome the limitations of the conventional CS-based approach under nonsparse SAR images. These contributions are verified through numerical simulations and experimental results.
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS
(2023)
Article
Geochemistry & Geophysics
Hanming Chen, Hui Zhou, Ying Rao
Summary: A novel FWI method is presented based on the fractional Laplacian viscoacoustic wave equation, with efficient gradient computation using the adjoint state method. The strategy utilizes Fourier pseudo-spectral time-domain numerical solutions for accurate representation of seismic wave behaviors. Wavefield reconstruction is achieved through reverse time-marching formulae, avoiding the need to store massive boundary values.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Lan Li, Meiping Song, Qiang Zhang, Yushuai Dong
Summary: This article proposes a global variation and local structure low-rank model (GLLR) for hyperspectral image (HSI) denoising by integrating spatial segmentation smoothing and spectral low-rank properties. The GLLR considers both global and local structure properties of HSIs, and effectively removes noise while retaining useful structural information in the HSI.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Optics
Shu-Feng Lin, Philippe Gentet, Di Wang, Seung-Hyun Lee, Eun-Soo Kim, Qiong-Hua Wang
Summary: The proposed full-color holographic three-dimensional display system utilizes an angular-compensating holographic optical element to achieve a simple structure for full-color display. This system eliminates the need for additional optical elements for synthesizing full-color reconstruction, leading to enhanced system integration and miniaturization.
OPTICS AND LASERS IN ENGINEERING
(2021)
Article
Geochemistry & Geophysics
Yikang Zheng, Yibo Wang
Summary: The traditional methods for GPR wavefield simulation suffer from numerical dispersion and complex geometry. We propose using PINNs to solve the GPR wave equation and model wavefield propagation. Deep fully connected networks approximate the equation's solution, and automatic differentiation with backpropagation is used for calculating partial derivatives. The PINNs solver for GPR wavefield simulation shows high accuracy and efficiency, and has potential for inverting electrical parameters from observed wavefields.
Article
Mathematics
Xinyu Pu, Baicheng Pan, Hangjun Che
Summary: Graph-based multi-view clustering aims to explore partition patterns using a similarity graph. Existing methods often lack information on the underlying low-dimensional space and struggle to handle noise in the graph. To address these issues, a novel method combining spectral embedding, non-convex low-rank approximation, and noise processing is proposed, outperforming other state-of-the-art methods on five datasets.
Article
Geochemistry & Geophysics
Shashin Sharan, Rongrong Wang, Felix J. Herrmann
GEOPHYSICAL JOURNAL INTERNATIONAL
(2019)
Article
Geochemistry & Geophysics
Bas Peters, Brendan R. Smithyman, Felix J. Herrmann
Article
Geochemistry & Geophysics
Philipp A. Witte, Mathias Louboutin, Navjot Kukreja, Fabio Luporini, Michael Lange, Gerard J. Gorman, Felix J. Herrmann
Article
Geochemistry & Geophysics
Curt Da Silva, Yiming Zhang, Rajiv Kumar, Felix J. Herrmann
Article
Computer Science, Software Engineering
Curt Da Silva, Felix Herrmann
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE
(2019)
Article
Geochemistry & Geophysics
Rajiv Kumar, Marie Graff, Ivan Vasconcelos, Felix J. Herrmann
GEOPHYSICAL PROSPECTING
(2019)
Article
Geochemistry & Geophysics
Philipp A. Witte, Mathias Louboutin, Fabio Luporini, Gerard J. Gorman, Felix J. Herrmann
Article
Computer Science, Interdisciplinary Applications
Rajiv Kumar, Bram Willemsen, Felix J. Herrmann, Alison Malcolm
COMPUTATIONAL GEOSCIENCES
(2019)
Article
Computer Science, Software Engineering
Fabio Luporini, Mathias Louboutin, Michael Lange, Navjot Kukreja, Philipp Witte, Jan Huckelheim, Charles Yount, Paul H. J. Kelly, Felix J. Herrmann, Gerard J. Gorman
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE
(2020)
Article
Geochemistry & Geophysics
Ali Siahkoohi, Gabrio Rizzuti, Felix J. Herrmann
Summary: This article utilizes Bayesian inference and deep neural networks to translate uncertainty in seismic imaging to uncertainty in tasks performed on the image, such as horizon tracking. A systematic approach is developed to address uncertainty due to noise in the data, characterizing it with a convolutional neural network and sampling from the posterior distribution to assess uncertainties.
Article
Geochemistry & Geophysics
Ali M. Alfaraj, D. J. (Eric) Verschuur, Felix J. Herrmann
Summary: This study proposes a low-rank-based residual statics estimation and correction method, which utilizes the low-rank structure of seismic data to estimate statics and overcomes the errors in conventional methods caused by near-surface effects and non-consistent effects. By performing low-rank approximation and crosscorrelation in the midpoint-offset-frequency domain, the method reduces the requirement for accurate rank selection and shares consistent statics across multiple frequency bands. Experimental results demonstrate a significant improvement in short-wavelength statics correction compared to conventional methods.
Article
Geochemistry & Geophysics
Ali Siahkoohi, Gabrio Rizzuti, Rafael Orozco, Felix J. Herrmann
Summary: Bayesian inference for high-dimensional inverse problems is computationally costly and requires selecting a suitable prior distribution. Amortized variational inference addresses these challenges by pretraining a neural network that acts as a surrogate conditional distribution. However, it relies on high-fidelity training data and is prone to errors if evaluated over data that are not from the training data distribution.
Article
Geochemistry & Geophysics
Yijun Zhang, Ziyi Yin, Oscar Lopez, Ali Siahkoohi, Mathias Louboutin, Rajiv Kumar, Felix J. Herrmann
Summary: Modern-day reservoir management and monitoring of geologic carbon storage require costly time-lapse seismic data collection. We demonstrate the use of graph theory techniques to optimize low-cost sparse 4D seismic data acquisition geometries. Our algorithm automatically produces sparse nonreplicated time-lapse acquisition geometries that favor wavefield recovery based on midpoint-offset-domain connectivity arguments.
Article
Computer Science, Artificial Intelligence
Emmanouil Daskalakis, Felix J. Herrmann, Rachel Kuske
SIAM JOURNAL ON IMAGING SCIENCES
(2020)
Article
Geosciences, Multidisciplinary
Mathias Louboutin, Michael Lange, Fabio Luporini, Navjot Kukreja, Philipp A. Witte, Felix J. Herrmann, Paulius Velesko, Gerard J. Gorman
GEOSCIENTIFIC MODEL DEVELOPMENT
(2019)