Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/231558
Title: LEARNING-BASED QUANTITATIVE IMAGING IN INVERSE PROBLEMS
Authors: ZHOU YULONG
ORCID iD:   orcid.org/0000-0001-9628-1828
Keywords: Learning-based, Quantitative Imaging, Near-field Scanning Microwave Microscopy, Inverse Scattering Problems, Subsurface Imaging, High Contrast
Issue Date: 1-Jun-2022
Citation: ZHOU YULONG (2022-06-01). LEARNING-BASED QUANTITATIVE IMAGING IN INVERSE PROBLEMS. ScholarBank@NUS Repository.
Abstract: The thesis applies learning-based quantitative imaging technologies to solve two inverse problems: the near field scanning microwave microscopy (NFSMM) and the inverse scattering problem (ISP). The original contributions are summarized as follows. Firstly, this thesis proposes a systematic material characterization method via the coaxial resonator based NFSMM (CR-NFSMM) with an arbitrary tip shape, including the modelling method and the inversion algorithm. Secondly, a learning-based method is proposed for the quantitative subsurface imaging via the CR-NFSMM in a non-destructive way. Numerical and experimental results show that the method could improve the image resolution and recover the dielectric property of the subsurface perturbation pixel-by-pixel. Thirdly, the modified contrast scheme (MCS) is proposed to tackle nonlinear inverse scattering problems. Numerical results show that MCS with the modified contrast input performs well in both 2D and 3D testing examples in real time after the offline training process, even in high relative permittivity cases.
URI: https://scholarbank.nus.edu.sg/handle/10635/231558
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
ZhouYL_Thesis.pdf3.86 MBAdobe PDF

OPEN

NoneView/Download

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.