Questions, comments and suggestions concerning VintaSoft Imaging .NET SDK.
Moderator: Alex
wittaya@Wac
Posts: 15 Joined: Wed Mar 06, 2019 7:54 am
Post
by wittaya@Wac » Wed Jul 10, 2019 6:57 am
Hi, Alex
Which "Vintasoft.Imaging.Ocr.RecognitionRegionType" I have to use because I found problem about OCR Thai language.
This is image file
p.s. I use Thai & English language for OCR
Best Regard
Wittaya WAC
Alex
Site Admin
Posts: 2397 Joined: Thu Jul 10, 2008 2:21 pm
Post
by Alex » Wed Jul 10, 2019 11:57 am
Hi Wittaya,
For understanding your problem we need to reproduce your problem on our side. Please send us (to
support@vintasoft.com ) a small working application, which allows to reproduce your problem.
Best regards, Alexander
wittaya@Wac
Posts: 15 Joined: Wed Mar 06, 2019 7:54 am
Post
by wittaya@Wac » Thu Jul 11, 2019 6:53 am
Hi, Alex
I fixed out myself by re-download tessdata for eng and tha,
its work now, but ocr with Thai and English correct around 70% both.
Best Regard
Wittaya WAC
Alex
Site Admin
Posts: 2397 Joined: Thu Jul 10, 2008 2:21 pm
Post
by Alex » Thu Jul 11, 2019 12:03 pm
Hi Wittaya,
its work now, but ocr with Thai and English correct around 70% both.
What resolution do your document images have? Tesseract OCR provides good text recognition results if document image has resolution 300 dpi or higher.
Best regards, Alexander
wittaya@Wac
Posts: 15 Joined: Wed Mar 06, 2019 7:54 am
Post
by wittaya@Wac » Thu Jul 11, 2019 12:57 pm
Hi, Alex
I use 200 dpi.
Best Regard
Wittaya WAC
Alex
Site Admin
Posts: 2397 Joined: Thu Jul 10, 2008 2:21 pm
Post
by Alex » Thu Jul 11, 2019 2:08 pm
wittaya@Wac wrote: Thu Jul 11, 2019 12:57 pm
I use 200 dpi.
Increase image resolution and I think you will have better OCR results.
Best regards, Alexander