Skip to content

Latest commit

 

History

History
39 lines (23 loc) · 2.01 KB

README.md

File metadata and controls

39 lines (23 loc) · 2.01 KB
Raw

Optical Character Recognition using Kohonen Map (self-organizing map (SOM)).

Project based on : Heaton, Jeff. Introduction to neural networks with java,2nd ed. Heaton Research. October,2008.[1]

According to Kohonen (1990, p.1) "...The Self-Organizing map has the special property of effectively creating spatially organized "internal representations" of various features of input signals and their abstractions of various features of input signals and their abstractions" ( Page 1)[2]

This project objective is to use the OCR with SOM project from the book and improve it to make it more didactic for students.And also able to run on android.

Picture 1 : Main Screen Picture 2: Adding Letter to the program
mainscreen.png AddLetter A.png
Picture 3: Neural Network (SOM) Trained Picture 4: Recognizing letter X
Trained.png RecognizeX.png
  • Quick summary
  • Version

This project is still on alpha version and still under development.

Who do I talk to?

Please contact the repository admins : Andre Lopes or Robin Stumm

#SPECIAL THANKS#

JEFF HEATON http://www.heatonresearch.com/ HEATON, JEFF. Introduction to Neural Networks with Java, Second Edition. 2 edition (October 1, 2008). Publish Location: Heaton Research, Inc, October 1, 2008. 440 pages.

References

1.Heaton, Jeff. Introduction to neural networks with java,2nd ed. Heaton Research. October,2008.

2.KOHONEN,Teuvo.The Self-Organizing Map.Retrieved 22 October,2014 from http://www.eicstes.org/EICSTES_PDF/PAPERS/The%20Self-Organizing%20Map%20(Kohonen).pdf