Quantitative transmission tomography for don-destructive imaging of stored grain and biological tissue
Loading...
Date
2021-08-25
Authors
Hughson, Max
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis presents the development, testing, and verification of transmission tomography software for the applications of small-scale biological imaging and large-scale stored-grain imaging.
Transmission tomography is a non-invasive technique for producing quantitative images of an object's physical properties.
This is done by interrogating the object with waves and measuring the resulting response at a set of measurement positions.
The properties of the received waves are analyzed, and used to calculate the properties of the object which is interrogated.
First, a mathematical formulation of transmission tomography is presented and explained.
The formulation is then used to build a numerical model of the physical properties which dictate wave transmission in two systems of linear algebraic equations.
The algorithms that are required for building the numerical model are then explained.
The rest of this thesis is devoted to a series of experiments.
These experiments show the usefulness of transmission tomography in some particular applications.
They also motivate the incremental development of features of the transmission tomography algorithm that was developed for this thesis.
The first experiment uses acoustic data to perform two-dimensional transmission tomography.
The first experiment shows that two-dimensional transmission tomography of acoustic data produces useful images of an object's wave speed.
The successful two-dimensional experiments motivate the two following experiments, which use electromagnetic data to perform three-dimensional transmission tomography.
In order to perform three-dimensional tomography of electromagnetic data, a modification is made to the numerical model.
The successful three-dimensional transmission tomography algorithm is then applied to track the volume of grain stored in a bin.
The final experiment uses acoustic data and the modified numerical model to perform three-dimensional transmission tomography to identify two distinct spherical objects which are submerged in a liquid medium.
Description
Keywords
Imaging, Basis, Matlab, Acoustic, Ultrasound, Electromagnetic, Grain, Agricultural, Time domain, Quantitative