Re: OCR question

[Duncan Watson [EMAIL PROTECTED]]

Excellent suggestion.  I found it.  I will see what I can do about some
edge detection to crop the images so that gocr has an easier time of
it.

Thanks
-Duncan

-- 
http://mail.python.org/mailman/listinfo/python-list

OCR question

[Duncan Watson[EMAIL PROTECTED]]

I run a small guild in an on-line MMO (City of Heroes).  As part of
this I maintain a roster of our membership and their level and various
in-game statistics.  Unfortunately I can't get the data from the MMO
directly.  I have been taking screenshots of the in-game roster and
then transcribing it into comma-delimited format for parsing by a
python script.

This works fine but some of our partner guilds are interested in my
roster.  So I was thinking I could try some limited OCR.  In this case
the font and image size will be known constants.  I would then try to
parse the screenshots directly.

Does anyone know of any libraries or toolkits I can use with this?  Or
should I just dig up some OCR texts and make a cheater version, quick,
dirty and a wee bit fragile.

I will be using PIL to read the images and convert to bmps/memory
images.  I just am curious where I can did up some OCR related data.

-Duncan

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: OCR question

[[EMAIL PROTECTED]]

try gocr?
command line ocr for linux
Not very pythonic, but I am unaware of any ocr implementations in
python.

It is gpl, so it should be to hard to write python wrappers if you
really want to do it right.

-- 
http://mail.python.org/mailman/listinfo/python-list