sikuli-driver team mailing list archive

Thread
Date

[Question #280467]: OCR custom fonts Training and Installation

To: sikuli-driver@xxxxxxxxxxxxxxxxxxx
From: Chathura <question280467@xxxxxxxxxxxxxxxxxxxxx>
Date: Mon, 04 Jan 2016 18:57:26 -0000
Reply-to: question280467@xxxxxxxxxxxxxxxxxxxxx
Sender: bounces@xxxxxxxxxxxxx

New question #280467 on Sikuli:
https://answers.launchpad.net/sikuli/+question/280467

I tried Sikuli OCR and text recognition is very poor to useless.

I used:
myregion=Region(t.x,t.y, 200, 200)
f=myregion.find("170")
f is emply meaning OCR text find fails to find 170 in custom myregion area.

 I wanted to include my own English fonts with Tesseract training. How would I go about creating my own training set and let Sikuli to use that training data for text recognition?
How to train and include custom Tesseract data sets?  I only need to train custom numbers for now.

-- 
You received this question notification because your team Sikuli Drivers
is an answer contact for Sikuli.