FontClustr
FontClustr copied to clipboard
Visually compares fonts, generates hierarchical clusters
FontClustr
FontClustr was conceived in January 2010 by Ian Katz. [email protected] / http://tinylittlelife.org
It uses computer vision algorithms to sort fonts by appearance.
FontClustr is free software licensed under the terms of the Apache 2.0 license.
Installation
FontClustr is written in Python 2 and JavaScript. It requires the following libraries:
Python
- fonttools
- opencv
- Python Imaging Library (or Pillow)
- numpy
JavaScript
Running npm install will fetch the following dependencies:
- d3
- markov-cluster
- minimist
Linux Support
Tested in Ubuntu 10.10, requiring the following packages: python-pygame python-opencv python-numpy python-imaging fonttools
Probably other packages, your feedback is appreciated.
OSX Support
You need to install opencv; I used the instructions here: https://jjyap.wordpress.com/2014/05/24/installing-opencv-2-4-9-on-mac-osx-with-python-support/
You need to pip install: PIL numpy fonttools
Usage
New Way
- Run
python fontclustr_json.py. This will generate a cache and a distance matrix, and save them for future use. - Run
node index.js report/distance_information_AaBbCcGgHhKkOoPpTtXx.json report/allClusters.json. This will calculate the clusters using a hacked approach to the Markov Clustering algorithm - Run
python -m SimpleHTTPServer 9999 - Open http://localhost:9999/report/ in a web browser
Old Way
Run FontClustr as "python fontclustr.py" or make fontclustr.py executable and run it directly.
There is experimental support for a GUI version of FontClustr called gfontclustr.py that caches results.
Right now, it sucks. Eventually it will be interactive. "python gfontclustr.py"