PDF version of this entire document
The current implementation overlays the image sequence with arrows based on a measure similar but not identical to http://en.wikipedia.org/wiki/Cross_correlationcross-correlation. It can position and initialise points based on several parameters and then show their movement in each frame transition. The next step will be to test this on a synthetic set and refine the algorithm as means of general improvement before validation experiments. This phase of the experiments deals with tracking but not any measurements (e.g. distances, stress) or statistical analysis.
Shown in the figures are some screen snapshots of the arrows put automatically on top of the image sets that are available to us. The next step will be considering the existing literature, then validation experiments with the synthetic data (tagged or not tagged). If within the synthetic dataset the tags are removed, then generating variation based on radii should be easier.
