PDF XChange Forum

Hi! Accidentally some letters were not recognized, which in my opinion are very legible: the word is for example DRAIN; BRAIN was recognized. So my 300 employee read misunderstandings using Your software and cant find repetitions using your search function: How often we can sell DRAIN in 200 pages? If Yout software often read BRAIN, we must look at every page to read every page with our eyes.

Thank You for Your very goot software,
Bernd

Hello Biber,

Thank you for your report.

Could you please tell us the exact version of your PDF Editor?

You can see that under the Main menu -> Help -> About.


Also, could you please tell us which OCR tool you have used
and if it possible please show us the settings
you have used during the recognition?


Regards.

Hello Biber,


Thank you for the provided information even if it is not everything I asked you for.


I made a few test and I will be grateful if you could tell me what part of this document is not correct:



Regards.

Hi! What do You mean with: which OCR-Tool? Is this Your OCR-tool, You asked for?
(see in post 1: ) ... the word is for example DRAIN; BRAIN was recognized. We want to know, how often we can sell DRAIN in for example 200 pages? Your software shows us, how often we can sell BRAIN.

Hello Biber,

I apologize for the delay.

What I meant was whether you have used the Default OCR or the Enhanced OCR tool.

I asked you to take a look at my test file, if there are the same mistakes made as in your tests, but after a close comparing with your file, I noticed that they are very similar.

I reported this problem to our developers, so, hopefully this problem will be resolved in the next version our the OCR tool.


Regards.

Thank You. We will wait. ...
The mistakes are the same (brain,
not drain).

Hello Biber7,

Thanks for the follow up!
We are still awaiting feedback from the people working on the OCR engine. As soon as we have any further news we will post here.

Kind regards,
Stefan

Hi all,

Today I bought the Enhanced OCR Version, too, because I expected better results.

The results are not even close to working with.

Thats the document:

That is the result:

ndspanl mil Ilppi. voll“-
besiehund aus 0rts-Nr.13 und 14


That was my setup:
Is there any way to improve the OCR significantly?

I am a big fan of your software anyway.

Kind regards,
Paul

It seems that you have tried Default OCR, not Enhanced one.

Find Enhance Scanned Pages command:

Hello RolleRolle,

Please try OVG's suggestion and see if that will help.

And if we may add some additional information for everyone:
We had been planning a release earlier this month which has been put back to mid December - one of the primary reasons for this is a significant improvement in OCR is expected with a free upgrad for all users with EOCR purchased and active maintenance at the time of the release. If all goes to plan that we are confident will improve this - Dramatically!

Kind regards,
Stefan

Biber7 wrote: ↑Wed Nov 11, 2020 3:57 pm Thank You. We will wait. ... The mistakes are the same (brain, not drain).

Remember: we will wait ...
Original: Al+G200S4
wrong OCR: Al+620084

... we will wait ...

Hello Biber7,

Is that latest image of yours from your original document in the first post?
I can't seem to find that in the original file.

Please update to V9 and try the new Enhanced OCR tool (it is now based on ABBYY's Fine Reader), as for me it gave a very decent result and I could not find any errors at my end with the OCR result. Kind regards,
Stefan

How to update to V9, please? Write a request to the IT-department to buy new licenses for 500 workstations?

Hello Biber7,

No need to buy anything at the moment! As you have active maintenance plan on your current license - all you need to do is download and install the new version - and it will directly work with the existing key!

Kind regards,
Stefan