Partial success on text searches - looking for suggestions on how to improve my success rate - Ubuntu tesseract 3.04.01 Sikulix

50 views
Skip to first unread message

Patrick Archibald

unread,
Feb 9, 2017, 12:32:28 AM2/9/17
to tesseract-ocr
Hi

Text search on my automation project is partially successful. Looking for advice on how to improve my results. Environment: Ubuntu 16.04, Tesseract 3.04.01, Java, Sikulix, rdesktop to Windows 2008 R2 server running MACC GUI. 

Successful find in the screenshot below with strings: History, Rewards, Custom, Comments, IA358, FTTH, 8431, Cred

Unsuccessful with strings: Credit, PATRICK, ARCHIBALD, IF050, Internet, 8431501000

I'm an OCR noob trying to use Sikulix and Tesseract to automate service orders via GUI software that doesn't have an API. Any tips appreciated. 

Thanks, PLA


Patrick Archibald

unread,
Feb 9, 2017, 8:32:33 AM2/9/17
to tesseract-ocr
jTessBoxEditor screenshot. Looks like tesseract doesn't like whatever font this GUI is using. Taking a look at the first screenshot in this thread, anybody know what font the GUI is using? 

Reply all
Reply to author
Forward
0 new messages