README_figures_analyzer.txt

Figures_analyzer version 1.0 15/03/2013

TOOL DESCRIPTION
--------------------
- It makes figure recognition through an HTM Cortical Learning Algorithm and a Bayesian analysis.

USAGE 
--------------------
- Write into the prompt "figures_analyzer_interface.py" and follow the graphical interface to perform the analysis OR 
- Write into the prompt "figures_analyzer_terminal folder1 folder2 folder2 foldern" where folder1,2,n are the names of
the folders which contain the figures to learn

- In both cases, two files called try_me1 and try_me2 are available to try the recognition mechanism. They contain images 
of the same object type but slightly different from the ones provided to the learning process. 

REQUIREMENTS
--------------------
- Have the following module installed: easygui (for the GUI version).
- Have inside the directory figures_analyzer several folders, each one with a kind of figures to learn
and whose name will be the same to the kind of figures it contains (i.e. folder square with different kind
of squares and folder cross with different kind of crosses). Several folders are provided inside the project
directory but additional ones could be add. 

LIMITATIONS
-------------------
- It is only able to work with a fixed size of matrices (20x20).
- Its performance is not good if the position of the objects in the image matrices vary

POSIBLE PROBLEM
-------------------
- If you have any problem when you use the program try this:  When you choose in the graphical interface the folders 
to learn try to click two times into each folder to choose it instead of click only once to select and another one in 
ok. 
   
copyright 2013 Adria Aterido Ballonga, Gerard Martinez Rossell and Bernardo Rodriguez Martin this software is distributed under the same terms as python.