Determination of the optimum metrics for the characterisation of scattering media

Housley, James (2010) Determination of the optimum metrics for the characterisation of scattering media. PhD thesis, University of Nottingham.

[img]
Preview
PDF (Thesis) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (2MB) | Preview

Abstract

Determining a scattering medium's absorption and scattering coefficients from measurements of the light reflected or transmitted from the medium is a common problem in various fields. The aim of this thesis was to calculate the errors in the determined coefficients using different combinations of light "metrics" such as reflectance and mean flight time under realistic noise conditions, as a result of which the optimum metric combination could be found.

This was investigated by the forward modelling of various metrics of detected light from a semi-infinite, homogeneous medium using the Diffusion Approximation. The normalised intensity and cumulants of the light's temporal point spread function (TPSF), were investigated as possible metrics and their form over a range of optical coefficients corresponding to in vivo human tissue described fully. These metrics were then used to provide simultaneous equations from which the medium's scattering and absorption coefficients could be calculated.

Errors in the metrics will propagate through to errors in the determined coefficients and a general method to calculate the extent of this propagation was described. To simulate realistic metric errors, a typical streak camera (Hamamatsu C5680) was modelled to determine its effect on the measurements of the metrics. Using this model, "error maps" showing the expected error in each metric's value over the range of absorption and scattering coefficients were produced. These were then applied to the general error analysis method. In full-field detection mode, it was shown that the combination of normalised intensity and first cumulant gave the most accurate answer for the medium's coefficients, while for spatially resolved detection, various combinations of reflectance and the mean time of flight were found to be the optimum metric pairs under different conditions.

Finally, a method of using the known characteristics (either from modelling or experiment) of a detector such as the streak camera to improve the accuracy of the determined coefficients was described.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Crowe, J.A.
Morgan, S.P.
Subjects: Q Science > QC Physics
Faculties/Schools: UK Campuses > Faculty of Engineering > Department of Electrical and Electronic Engineering
Item ID: 11356
Depositing User: EP, Services
Date Deposited: 03 Nov 2010 09:06
Last Modified: 19 Oct 2017 11:40
URI: https://eprints.nottingham.ac.uk/id/eprint/11356

Actions (Archive Staff Only)

Edit View Edit View