A phase unwrapping algorithm based on Branch cuts for living cell's interference pattern

Fibroblast is the main part in the loose connective tissue and differentiates from the mesenchymal cell when it is in embryo. It exhibits highly reproducible growth kinetics and reproducible healing dynamics in the scratch-wound assay and the height of it could show this prediction. In order to measure the height of these cells, we construct an interferometer measuration system. As we all know, the interference pattern should be unwrapped first, there are plenty of methods that are under research. In this paper we want to find out a typical methods that could be used in living cell's interference pattern during image processing, and also we can get the conclusion that how to use the method and why it is fit to unwrap the phase of cells. There are mainly three parts in this paper: Firstly, we have designed an Interference system which can be used to get the interference pattern, here we used multiphase interference microscope to measure the cell height. Secondly, a typical method which is based on Goldstein's branch cuts algorithm were used to guide the way that how the phase is unwrapped, this method is the most efficient way to phase unwrapping, and it could induct the unwrapping path through using the branch cut method which could get rid of the residues as much as it could be. As a comparison, we also used some other methods to find different results. Such as the quality-guided path following phase unwrapping; and the Costantini phase unwrapping. Finally, we analyzed the results of the three-dimensional model of the cell surface topography, as a result of the various noises during the experiment, all these unwrapping methods above can't eliminate all the residues and noises, but compared with the other results, the Goldstein's branch cut method has the fittest advantages, it gives the most fluent topography of the living cells.