中文

Multi-parameter Full-waveform Inversion of 3D On- ground GPR with a Modified Total Variation Regularization Scheme

Hits:

  • Release time:2020-08-07

  • Impact Factor:3.534

  • DOI number:10.1109/LGRS.2020.2976146

  • Affiliation of Author(s):School of Geosciences and Info-Physics, Central South University, Changsha 410083, China

  • Teaching and Research Group:Applied Geophysics

  • Journal:IEEE Geoscience and Remote Sensing Letters

  • Place of Publication:UNITED STATES

  • Funded by:41574116; 41774132; CX2017B052

  • Key Words:Edge-based finite element method (EFEM);frequency domain, full-waveform inversion (FWI);ground penetrating radar (GPR);limited-memory Broyden-Fletcher-Goldfarb-Shanno(L-BFGS) method;multiparameter

  • Abstract:Ground-penetrating radar (GPR) full-waveform inversion (FWI) is an emerging near-surface high-resolution imaging technology that has been widely used in GPR 2-D imaging to better quantitatively evaluate the permittivity and conductivity of the subsurface. However, the 2-D imaging method limits the applicability and accuracy of FWI for the 3-D object cases. In this letter, we develop a 3-D FWI algorithm involving permittivity and conductivity for GPR in the frequency domain. The method employs the edge-based finite element method to solve the forward problem and the limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) method to solve the inverse problem, which avoids the calculation of the Hessian matrix. In particular, by using the parameter scaling and scaling factor, it is convenient to carry out the simultaneous inversion of permittivity and conductivity with high efficiency. In addition, a modified total variation regularization scheme is utilized for ensuring the stability of the inversion, as well as identifying the abnormal body boundary more effectively. As a demonstration, a synthetic 3-D model experiment is presented to test the proposed algorithm. The results show that the proposed inversion strategy can accurately reconstruct the permittivity structure together with the conductivity distribution from the 3-D on-ground GPR data.

  • First Author:王珣

  • Indexed by:Applied Research

  • Correspondence Author:冯德山

  • Document Code:DOI:10.1109/LGRS.2020.2976146

  • Discipline:地质资源与地质工程

  • Document Type:J

  • ISSN No.:1545-598X

  • Translation or Not:no

  • Date of Publication:2020-03-05

  • Included Journals:SCI


  • Zip Code:

  • Fax:

  • Postal Address:

  • Office Phone:

  • Mobile:

  • Email:

Central South University  All rights reserved  湘ICP备05005659号-1 Click:
  MOBILE Version

The Last Update Time:..