Parametric inversion of experimental electromagnetic imaging data using synthetically-trained neural networks

Simple item page
dc.contributor.authorEdwards, Keeley
dc.contributor.examiningcommitteeHenry, Christopher (Electrical and Computer Engineering) Pistorius, Stephen (Physics and Astronomy)en_US
dc.contributor.supervisorJeffrey, Ian (Electrical and Computer Engineering) Gilmore, Colin (Electrical and Computer Engineering)en_US
dc.date.accessioned2021-07-27T21:13:10Z
dc.date.available2021-07-27T21:13:10Z
dc.date.copyright2021-07-22
dc.date.issued2021en_US
dc.date.submitted2021-07-20T19:04:40Zen_US
dc.date.submitted2021-07-22T21:37:51Zen_US
dc.degree.disciplineElectrical and Computer Engineeringen_US
dc.degree.levelMaster of Science (M.Sc.)en_US
dc.description.abstractElectromagnetic imaging is a method that aims to reconstruct a spatial map of the material properties of a target through non-invasive measurements, allowing an observer to see inside objects, such as the human body, geological formations or large containers, without disturbing or entering them. This thesis evaluates the use of synthetically trained neural networks for performing phaseless parametric inversion of experimental electromagnetic imaging data. Fully-connected neural networks are used as a tool to obtain bulk parameters describing the geometry and complex-valued permittivity of materials within the region of interest (ROI) from experimental data. This data can subsequently be used for calibration and as prior information to facilitate full-phase 3D inversion of the data. These networks are applied to two different applications, one in agriculture and one in biomedical imaging, demonstrating that this method can be generalized to imaging systems at significantly different scales. In the agricultural application, synthetically-trained neural networks are used to infer bulk parameters from raw, uncalibrated experimental data for two different grain storage bins. This presents a promising tool for calibrating data collected from uncooperative systems where conventional calibration methods are not practical. For commercial grain monitoring systems, where multiple measurements are taken daily, neural networks present a cost-effective long-term alternative to existing iterative parametric inversion methods. For the biomedical application, a two-stage workflow is presented, demonstrating the suitability of the synthetically trained machine-learning-based parametric inversion to obtain the prior information needed for a second stage inversion using the Contrast Source Inversion (CSI) method. Noisy synthetic data representing a model of a human breast is used to simulate experimental data in the two-stage workflow, and preliminary bulk parametric inversion (neural network) results are given for experimental microwave breast imaging data. The results presented herein demonstrate that synthetically-trained neural networks can successfully infer bulk parameters from noisy synthetic and experimental data. Synthetic experiments for the microwave breast imaging case show that the inferred parameters are of sufficient quality to be used as prior information for full phase inversion and 3D image reconstruction.en_US
dc.description.noteOctober 2021en_US
dc.identifier.citationK. Edwards, V. Khoshdel, M. Asefi, J. LoVetri, C. Gilmore, and I. Jeffrey, “A machine learning workflow for tumour detection in breasts using 3d microwave imaging,” Electronics, vol. 10, no. 6, 2021. [Online]. Available: https://www.mdpi.com/2079-9292/10/6/674en_US
dc.identifier.citationK. Edwards, K. Krakalovich, R. Kruk, V. Khoshdel, J. LoVetri, C. Gilmore, and I. Jeffrey, “The implementation of neural networks for phaseless parametric inversion,” in 2020 XXXIIIrd General Assembly and Scientific Symposium of the International Union of Radio Science. IEEE, 2020, pp. 1–3.en_US
dc.identifier.citationK. Edwards, V. Khoshdel, M. Asefi, J. LoVetri, C. Gilmore, and I. Jeffrey, “Recovery of prior information for breast microwave imaging using neural networks,” in Accepted to URSI GASS 2021, 2021.en_US
dc.identifier.citationK. Edwards, N. Geddert, K. Krakalovich, R. Kruk, M. Asefi, J. Lovetri, C. Gilmore, and I. Jeffrey, “Stored grain inventory management using neural-network-based parametric electromagnetic inversion,” IEEE Access, vol. 8, pp. 207 182–207 192, 2020.en_US
dc.identifier.citationK. Edwards, J. LoVetri, C. Gilmore, and I. Jeffrey, “A machine learning method for characterization of complex grain air interfaces in grain storage bins,” Accepted to ANTEM 2021, 2021en_US
dc.identifier.urihttp://hdl.handle.net/1993/35756
dc.language.isoengen_US
dc.rightsopen accessen_US
dc.subjectElectromagnetic imagingen_US
dc.subjectParametric inversionen_US
dc.subjectMachine learningen_US
dc.subjectElectromagnetic inverse scatteringen_US
dc.subjectCalibrationen_US
dc.subjectPhaseless imagingen_US
dc.titleParametric inversion of experimental electromagnetic imaging data using synthetically-trained neural networksen_US
dc.typemaster thesisen_US
Files
Original bundle
Now showing 1 - 2 of 2
Thumbnail Image
Name:
IEEE_Permission.pdf
Size:
90.78 KB
Format:
Adobe Portable Document Format
Description:
IEEE Permission
Download
Thumbnail Image
Name:
Edwards_Keeley.pdf
Size:
9.51 MB
Format:
Adobe Portable Document Format
Description:
Thesis
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
2.2 KB
Format:
Item-specific license agreed to upon submission
Description:
Download
Collections
FGS - Electronic Theses and Practica