We begin with an introduction to the field, covering some background information that is useful in understanding the issues. Other research in the area is then reviewed.
The hardware and software requirements are considered, and the choices made are then justified.
The software developed for this thesis is summarised and documented, and the techniques employed for data collection are explained.
The different techninques used in analysing the data and the techniques eventually employed are then explained, illustrated, tested and assessed for their accuracy.
These features are then put together and tested. Special attention paid to the potentials for expansion of the techniques to larger lexicon (i.e. greater number of signs) systems.
The conclusions we can draw from this work are documented, and a considerable number of avenues for future research are suggested.
I'll try to keep it as interesting as possible, and I hope you don't get bored.