Mark_in_Hollywood
October 16th, 2008, 08:54 PM
While scanning the Tesseract forum at Google Groups, I read:
http://groups.google.com/group/tesseract-ocr/browse_thread/thread/58ca6a1aba787133?hl=en
Filip Dominec
View profile More options Oct 9, 10:57 am
From: Filip Dominec <filip.domi...@gmail.com>
Date: Thu, 9 Oct 2008 10:57:18 -0700 (PDT)
Local: Thurs, Oct 9 2008 10:57 am
Subject: Tesseract GUI
Hi, as I found Tesseract to be the probably best free OCR available, I
decided to write "gui-tesseract" - a little graphical frontend in
python.
Planned features:
1) Batch processing over all *.jpg, *.jpeg images in selected
directory.
2) Optionally crop, rotate, normalize (etc?) the image using
imagemagick. (The crop region can be visually chosen on the image
preview, as the gui-tesseract expects the images were scanned/
photographed so that the text regions are on the same position.)
3) Simple interface - in case of well-prepared images, it only takes
one click to process them.
4) Automatic numbering of output pages (all/even/odd). Tool to merge
multiple *.txt pages into one (yes, using cat).
I expect the code not to exceed 20 kB. Would it be possible to
distribute this script with Tesseract? It could provide effective OCR
function for many (not only newbie) Linux users.
Can someone (with a whole lot more Know-How that I have contact this guy and help him?
http://groups.google.com/group/tesseract-ocr/browse_thread/thread/58ca6a1aba787133?hl=en
Filip Dominec
View profile More options Oct 9, 10:57 am
From: Filip Dominec <filip.domi...@gmail.com>
Date: Thu, 9 Oct 2008 10:57:18 -0700 (PDT)
Local: Thurs, Oct 9 2008 10:57 am
Subject: Tesseract GUI
Hi, as I found Tesseract to be the probably best free OCR available, I
decided to write "gui-tesseract" - a little graphical frontend in
python.
Planned features:
1) Batch processing over all *.jpg, *.jpeg images in selected
directory.
2) Optionally crop, rotate, normalize (etc?) the image using
imagemagick. (The crop region can be visually chosen on the image
preview, as the gui-tesseract expects the images were scanned/
photographed so that the text regions are on the same position.)
3) Simple interface - in case of well-prepared images, it only takes
one click to process them.
4) Automatic numbering of output pages (all/even/odd). Tool to merge
multiple *.txt pages into one (yes, using cat).
I expect the code not to exceed 20 kB. Would it be possible to
distribute this script with Tesseract? It could provide effective OCR
function for many (not only newbie) Linux users.
Can someone (with a whole lot more Know-How that I have contact this guy and help him?