Filters








423 Hits in 5.2 sec

Estimating flow parameter distributions using ground-penetrating radar and hydrological measurements during transient flow in the vadose zone

Michael B. Kowalsky, Stefan Finsterle, Yoram Rubin
2004 Advances in Water Resources  
Through a synthetic example, performance of the method is evaluated under various conditions, and some conclusions are made regarding the joint use of transient GPR and hydrological measurements in estimating  ...  In the present work, an inverse technique is presented which allows for the estimation of flow parameter distributions and the prediction of flow phenomena using GPR and hydrological measurements collected  ...  [4] , who investigated the use of ERT and crosshole GPR, collected during a tracer injection test, to estimate by trial and error the effective value of saturated hydraulic conductivity.  ... 
doi:10.1016/j.advwatres.2004.03.003 fatcat:4qhwsalkmbc6nm6o6pzlrruxem

Estimating flow parameter distributions using ground-penetrating radar and hydrological measurements during transient flow in the vadose zone

M KOWALSKY
2004 Advances in Water Resources  
Through a synthetic example, performance of the method is evaluated under various conditions, and some conclusions are made regarding the joint use of transient GPR and hydrological measurements in estimating  ...  In the present work, an inverse technique is presented which allows for the estimation of flow parameter distributions and the prediction of flow phenomena using GPR and hydrological measurements collected  ...  [4] , who investigated the use of ERT and crosshole GPR, collected during a tracer injection test, to estimate by trial and error the effective value of saturated hydraulic conductivity.  ... 
doi:10.1016/s0309-1708(04)00045-4 fatcat:rghpvmtw6za6lik5uj7dzprveq

Deep Learning-based Rebar Clutters Removal and Defect Echoes Enhancement in GPR Images

Jing Wang, Kefu Chen, Hanchi Liu, Jiaqi Zhang, Wenqiang Kang, Shufan Li, Peng Jiang, Qingmei Sui, Zhengfang Wang
2021 IEEE Access  
In this study, a deep learning-based method for rebar clutters removal and defect echoes enhancement in GPR B-scan images was proposed.  ...  The residual-inception blocks and attention modules were designed in the network as per the characteristics of the task.  ...  [26] designed a deep neural network for the relative permittivity inversion from GPR data. Liu et al.  ... 
doi:10.1109/access.2021.3088630 fatcat:g6qatv2q5ng3bauvtnuihrgupa

RCE-GAN: A Rebar Clutter Elimination Network to Improve Tunnel Lining Void Detection from GPR Images

Yuanzheng Wang, Hui Qin, Yu Tang, Donghao Zhang, Donghui Yang, Chunxu Qu, Tiesuo Geng
2022 Remote Sensing  
Ground penetrating radar (GPR) is one of the most recommended tools for routine inspection of tunnel linings.  ...  The designed network has two sets of generators and discriminators, and by introducing the cycle-consistency loss, the network is capable of learning high-level features between unpaired GPR images.  ...  The novel contributions of the proposed method are as follows: (1) Designing a deep learning network (RCE-GAN) to eliminate rebar clutters in GPR images. (2) Two sets of generators and discriminators are  ... 
doi:10.3390/rs14020251 fatcat:i2gx6gizindapi6xhgqs5ubone

Efficient Design Of Peptide-Binding Polymers Using Active Learning Approaches [article]

Assima Rakhimbekova, Anton Lopukhov, Natalia L. Klyachko, Alexander Kabanov, Timur I. Madzhidov, Alexander Tropsha
2021 bioRxiv   pre-print
Active learning (AL) has become a subject of active recent research both in industry and academia as an efficient approach for rapid design and discovery of novel chemicals, materials, and polymers.  ...  We have investigated the dependency of AL performance on the size of the initial training set, the relative complexity of the task, and the choice of the initial training dataset.  ...  Y-MAX (deep and light green) and Y-VAR (deep and light blue) strategies were tested. The initial set contains 10 points, 5 objects were added at every AL cycle.  ... 
doi:10.1101/2021.12.17.473241 fatcat:hgxup7jo6fhdxpnxybsermfn7e

Hybrid computational–experimental data-driven design of self-assembling π-conjugated peptides

Kirill Shmilovich, Sayak Subhra Panda, Anna Stouffer, John D. Tovar, Andrew L. Ferguson
2022 Digital Discovery  
A hybrid computational–experimental active learning workflow efficiently discovers π-conjugated peptides with superior capabilities for programmed self-assembly into pseudo-1D nanofibers.  ...  Acknowledgements Notes and references  ...  The predictions of the two GPRs are passed to a multi-objective BO routine that seeks to simultaneously maximize k and R g using the method of random scalarizations. 39, 40 The trained GPRs for k and  ... 
doi:10.1039/d1dd00047k fatcat:w7lp3z3rxfg6fmwcma5ejqf35q

Deep active subspaces - a scalable method for high-dimensional uncertainty propagation [article]

Rohit Tripathy, Ilias Bilionis
2019 arXiv   pre-print
A problem of considerable importance within the field of uncertainty quantification (UQ) is the development of efficient methods for the construction of accurate surrogate models.  ...  active subspace projection matrix, and couple this formulation with deep neural networks.  ...  We allow for the possibility that our measurement from the computer code may be noisy, i.e., y = f (ξ ) + ε, where ε is a random variable (the measurement noise might arise as a consequence of quasi-random  ... 
arXiv:1902.10527v2 fatcat:26kt4svukjbrljlvbid6cqmy3a

Machine learning pipeline for battery state of health estimation [article]

Darius Roman, Saurabh Saxena, Valentin Robu, Michael Pecht, David Flynn
2021 arXiv   pre-print
In this paper, we design and evaluate a machine learning pipeline for estimation of battery capacity fade - a metric of battery health - on 179 cells cycled under various conditions.  ...  This work provides insights into the design of scalable data-driven models for battery SOH estimation, emphasising the value of confidence bounds around the prediction.  ...  The paper explores four algorithms: Bayesian ridge regression (BRR), Gaussian process regression (GPR), random forest (RF), and a deep ensemble of neural networks (dNNe), as the base algorithm for BHUMP  ... 
arXiv:2102.00837v1 fatcat:rcwl2totgfhx3f5f5stfma2nrq

Discovery of Self-Assembling π-Conjugated Peptides by Active Learning-Directed Coarse-Grained Molecular Simulation [article]

Kirill Shmilovich, Rachael A. Mansbach, Hythem Sidky, Olivia E. Dunne, Sayak Subhra Panda, John D. Tovar, Andrew L. Ferguson
2020 arXiv   pre-print
This work establishes new understanding of DXXX-OPV3-XXXD assembly, identifies promising new candidates for experimental testing, and presents a computational design platform that can be generically extended  ...  capable of assembling pseudo-1D nanoaggregates with good stacking of the electronically-active π-cores.  ...  candidate space for experimental synthesis and testing.  ... 
arXiv:2002.01563v1 fatcat:uvxbwpabmbg6pobgdr5xx5uaiu

Dictionary Learning for Adaptive GPR Landmine Classification [article]

Fabio Giovanneschi, Kumar Vijay Mishra, Maria Antonia Gonzalez-Huici, Yonina C. Eldar, Joachim H. G. Ender
2019 arXiv   pre-print
We use a Kolmogorov-Smirnoff test distance and the Dvoretzky-Kiefer-Wolfowitz inequality for the selection of DL input parameters leading to enhanced classification results.  ...  Ground penetrating radar (GPR) target detection and classification is a challenging task.  ...  ACKNOWLEDGEMENTS The authors acknowledge valuable assistance from David Mateos-Núñez for Section VI-D.  ... 
arXiv:1806.04599v2 fatcat:5dy22ktfpbeffb4sxbxtis5gxm

Unsupervised Image Regression for Heterogeneous Change Detection [article]

Luigi T. Luppino, Filippo M. Bianchi, Gabriele Moser, Stian N. Anfinsen
2019 arXiv   pre-print
With the identified pixels as pseudo-training data, we learn a transformation to map the first image to the domain of the other image, and vice versa.  ...  Notably, the random forest regression approach excels by achieving similar accuracy as the other methods, but with a significantly lower computational cost and with fast and robust tuning of hyperparameters  ...  In the learning phase, random pairs of co-registered image patches are encoded with a single code, which jointly indexes both dictionaries.  ... 
arXiv:1909.05948v1 fatcat:qkdti4xkaveslm5un4cw7r5yxa

Electrical imaging for localizing historical tunnels at an urban environment

Ana Osella, Patricia Martinelli, Vivian Grunhut, Matías de la Vega, Néstor Bonomo, Marcelo Weissel
2015 Journal of Geophysics and Engineering  
Acknowledgments This work was partially supported by CONICET and ANPCyT-Argentina.  ...  , GPR and ERT to define subsurface characteristics and search for possible structural damage in the subfloor of a church).  ...  We performed different tests and found that, in general, good electrode contact and a stable injection of current could be achieved by burying the electrodes at least 15 cm deep, to overcome the cobblestone  ... 
doi:10.1088/1742-2132/12/4/674 fatcat:v3fsid7tobh6va7mkeu2wqmh7u

A Human Visual System Inspired No-Reference Image Quality Assessment Method Based on Local Feature Descriptors

Domonkos Varga
2022 Sensors  
Objective quality assessment of natural images plays a key role in many fields related to imaging and sensor technology.  ...  To prove the efficiency of the proposed method, it was compared to 16 state-of-the-art NR-IQA techniques on five large benchmark databases, i.e., CLIVE, KonIQ-10k, SPAQ, TID2013, and KADID-10k.  ...  Acknowledgments: We thank the academic editor and the anonymous reviewers for their careful reading of our manuscript and their many insightful comments and suggestions.  ... 
doi:10.3390/s22186775 pmid:36146123 fatcat:c2rlz3e3hfc3fm23xvwoitgtvu

Mobile Systems from a Validation Perspective: a Case Study

Helene Waeselynck, Zoltan Micskei, Minh Duc Nguyen, Nicolas Riviere
2007 Sixth International Symposium on Parallel and Distributed Computing (ISPDC'07)  
The protocol has been analyzed by reviewing the specification and the code, and then by testing the implementation. The outcomes provides us with hints for research direction.  ...  Advances in wireless networking have yielded the development of mobile applications. However, sound technology to specify, design and validate such applications is still to be investigated.  ...  The UML models proved a useful support for the analysis. They allowed us to gain deep insight into the code executed by each node.  ... 
doi:10.1109/ispdc.2007.37 dblp:conf/ispdc/WaeselynckMNR07 fatcat:fwejuaqe2rbahnvqven5ib6mjm

Wavelet Scattering Network-Based Machine Learning for Ground Penetrating Radar Imaging: Application in Pipeline Identification

Yang Jin, Yunling Duan
2020 Remote Sensing  
A double-channel framework is designed with wavelet scattering networks followed by support vector machines to determine the existence of pipelines on vertical and horizontal traces separately.  ...  Pipeline locations and diameters are convenient to determine from the reconstructed profiles of both simulated and practical GPR signals.  ...  Introduction Ground penetrating radar (GPR) is a well established non-destructive technology for the geophysical investigation of subterranean structures and substances by propagating electromagnetic waves  ... 
doi:10.3390/rs12213655 fatcat:jtvds5mcf5bp5jmyaxni3zbcy4
« Previous Showing results 1 — 15 out of 423 results