An image processing procedure to perform a quasi-real time dimensional analysis of
blood vessels undergoing morphological changes in a perfusion myograph is
developed. Images of blood vessels are recorded using perfusion myography and
video microscopy. The vessels' edges are detected automatically using an edge-tracking
algorithm which combines a tracking method with image processing
techniques, including: gradient operators and threshold. Because of the scattering
effects that occur when the images are recorded, the edges found by the algorithm are
not the actual vessels' edges. An attempt to correct this effect using a low order optical
model based on Beer-Lambert's law of light transmission through absorbing media is
proposed and tested.
A master's thesis submitted in partial fulfilment of the requirements for the award of Master of Philosophy at Loughborough University.