I wrote a python script using the OpenCV framework to detect digits on a credit or insurance card.
How does it work?
The code works simply as two parts. The aim of the first part is to train the script with possible images. Once the training is done, you can test your scanned image (i.e. credit card or insurance card).
Sample recognition output?
Here you can see the output of a successful digits recognition:
Demo training part
Simply tap on the keyboard key that is displayed in a red rectangle
You can get access to the source code via Github: