3

According to How can instantaneously extract text from a screen area using OCR tools? and system responses I use 

sh screen_ts.sh
, the content of which is 
#!/bin/bash
scrot -s zkoSCR_IMG.png
tesseract zkoSCR_IMG.png zkoSCR
cat zkoSCR.txt|xclip -i -selection clipboard
. The Terminal responses 
Tesseract Open Source OCR Engine v3.02 with Leptonica
, but the text is recognized with too many errors, even in the Terminal window itself. The ABBYY Screenshot Reader in Windows XP is far better. Can I achieve such better results in the Ubuntu? How? Thanks.
Community
  • 1
Tomáš Pečený
  • 1,407
  • 7
  • 19
  • 45
  • 1
    Have a look at http://stackoverflow.com/questions/11489824/how-do-i-choose-between-tesseract-and-opencv for a discussion of the subject. – guntbert Sep 07 '13 at 21:39
  • 1
    Hey Tomas. I edited the answer in the Q&A you linked and updated it with an enhanced script. The detection rate should now be much better. Hope it works for you. I will flag this question as a duplicate because I think both Q&As cover the exact same issue. – Glutanimate Sep 13 '13 at 16:14

0 Answers0