https://github.com/garethgeorge/cs1l-pythonocr

A project created in 2015Q1 for CS1L. It takes an image file as an input and returns the text that it recognizes in the image.
https://github.com/garethgeorge/cs1l-pythonocr

Last synced: about 1 month ago
JSON representation

A project created in 2015Q1 for CS1L. It takes an image file as an input and returns the text that it recognizes in the image.

• Host: GitHub
• URL: https://github.com/garethgeorge/cs1l-pythonocr
• Owner: garethgeorge
• Created: 2015-12-01T12:13:14.000Z (about 9 years ago)
• Default Branch: master
• Last Pushed: 2015-12-02T22:37:47.000Z (about 9 years ago)
• Last Synced: 2023-02-26T20:56:32.583Z (almost 2 years ago)
• Language: Python
• Size: 445 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# PYTHON OCR

by Gareth George

Python OCR is a basic optical character recognition library written in python. It takes an image file as it's input and returns what it believes the contents of the image file to be.

It uses a neural network (using pybrain's implementation) to perform the actual digit classification. This neural network is trained on samples of a number of fonts so it can recognise various typefaces. By default Ariel and Helvetica are coded in though more can be added in lib fontydatasets.py.

### THE STEPS

 - simple thresholding and image binarization

 - letter seperation via a flood fill type algorithm

 - image resizing to 28x28 pixels.

 - classification with pybrain neural network

# INSTALLATION

DEPENDENCIES

 - pybrain

 - python

 - PIL