Skip to content

GeorgeSaman/Optical-Character-Recognition-BackPropagation

Repository files navigation

CharacterRecognition-BackPropagation

Character Recognition using an ANN.

##How To use.

  • Open OCR.py, this is the main script.
  • Navigate to Parameters section, this is where you can adjust all input parameters to your network, for example,
    1. Number of hidden neurons
    2. learning Rate
    3. Momentum
    4. Size of the normalized image.
  • Samples folder contain all training samples for OCR-EXTENDED font.
  • Paragraphs contain the input document(IMAGE) to be recognized. To input your own document, change the documentLocation variable to your document name.
  • CharacterRecognition file contains a three layer neural network trained using Back Propagation.

For further read, take a look at the pdf file provided with the code.

About

Optical Character Recognition using An Artificial Neural Network Trained Using Back Propagation.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages