inconsistent ocr result with the ssame page

Questions, comments and suggestions concerning VintaSoft Imaging .NET SDK.

Moderator: Alex

Post Reply
Posts: 3
Joined: Tue Jan 14, 2020 1:06 pm

inconsistent ocr result with the ssame page

Post by Villersfr »

hi Support,
hi Community,

We are facing a very strange behavior of the ocr module.

When we apply the OCR 2 times on exactly the same original, the results are not the same (globally it is always good) but for example, the first time it will not ocr 2 or 3 words and the second time it ocr the 2 or 3 previous missing words but does not ocr 2 or 3 other words !!!

When we use the vintasoft demo provided (on line or build on our server) the résult are simply perfect and ALL the words are detected.

Of course, we are using exactly the same settings and original for the test comparison, we also repeat the tests at different resolution but the problem persist.

Our process is as follow (scan from a Multifonctionnal printer at 300dpi, generatibg a pdf image) - we apply the document cleanup (with the vintasoft module) - we apply ocr (with vintasoft tesseract ocr module) - we generate the searchable pdf (with the vintasoft pdf module).

Any idee ?

Best regards

Site Admin
Posts: 2045
Joined: Thu Jul 10, 2008 2:21 pm

Re: inconsistent ocr result with the ssame page

Post by Alex »

Hi Frederic,

We already answered to your question in our support center.

Best regards, Alexander
Post Reply