Retro Color Analysis

By FireElementalNE

This program is made to analyze old game maps for color information.
In reality you can analyze any set of images.
retro_analysis.py is the main program, you can split the input, output and HTML
generation steps via the command line arguments.

usage: retro_analysis.py [-h] [-i INPUT] [-t]

Old game color analyzer

optional arguments:
  -h, --help            show this help message and exit
  -i INPUT, --input INPUT
                        Input Folder (individual game)
  -t, --threaded        thread the execution

NOTE: The -t option does nothing (work in progress)

main.py is a wrapper used for convince and simplicity.

usage: main.py [-h] -g GAME

Wrapper function

optional arguments:
  -h, --help            show this help message and exit
  -g GAME, --game GAME  Game name

Input

For the input stage the given file directory is searched.
This directory given needs to be within the maps folder.
Each PNG Image within the input subdirectory of this folder
is analyzed for color information. The 10 most common colors are selected
(which is configurable in global_values.py along with many other variables) and
made into a color swatch.

Three statistics are also calculated:

1. The average distance between the top colors in HSV space
2. The average distance between the top colors in LAB space
3. The average distance between the top colors in RGB space

A Dendrogram of the top 40 colors (which is also configured in global_values.py)
is also made. The distance matrix that is used to create the Dendrogram can be created using the values
of the colors in RGB, HSV or LAB space (this is configurable in global_values.py).

Output

Outputs from the input phase are taken as inputs to the output phase creating
an aggregate result for each game folder.

HTML

The HTML builder takes all of the input and output data and makes a nice pretty
bootstrap based web page on the fly. It holds all of the information discussed
above.

External Dependencies

• Python Imaging Library found Here.
• SciPy found here.
• Numpy found here.
• Matplotlib found here.

Notes

make_test_images.py is used to create test images to upload to Github.

usage: make_test_images.py [-h] [-n NUMBER]

Create Test data

optional arguments:
  -h, --help            show this help message and exit
  -n NUMBER, --number NUMBER
                        number of test files to create

Program written in Python 2.7.6