0

I need to automate a extract of info from a Flash App. My idea was to take a screenshot and get the text from there, but using Pytesser isn't working (it returns nothing, don't know if it's because i'm doing something wrong or because it's too small).

The text is black arial size 8 text in a white background, no noise, and the text is pretty simple (mostly things like "9:32" or "Cristiano Ronaldo"). How would you face this problem in a simple way and using Python?

Just in case it's needed, i'm using Windows XP and Python 2.7

Thank you in advance.

user2376733
  • 13
  • 3
  • Have you any code on how you extract the screenshot? or is that an external application doing the work, or even manually? – Torxed May 13 '13 at 07:41
  • This might help you: http://stackoverflow.com/questions/5799946/python-ocr-module-in-linux – Torxed May 13 '13 at 07:41
  • I'm using PIL to take the screenshot and grab the interesting part. I'll take a look to that post, thanks! – user2376733 May 13 '13 at 08:05
  • If the text is too small, try 'blowing up' the pixels, for example 2x or 3x. That should produce sufficient font size for any OCR to read. However, 8 points is usually OK for a modern OCR (not sure about the one you are using). – Ilya Evdokimov May 14 '13 at 00:33

0 Answers0