1 | initial version |
How do you want to help? By implementing an OCR algorithm and submit a pull request? Many example of that in OpenCV exist so I guess it is just filtering out the basics and create a stable technique.
Some interesting links: