Multi-language OCR can take a long time  SOLVED

Forum for the PDF-XChange Editor - Free and Licensed Versions

Moderators: TrackerSupp-Daniel, Tracker Support, Paul - Tracker Supp, Vasyl-Tracker Dev Team, Chris - Tracker Supp, Sean - Tracker, Ivan - Tracker Software, Tracker Supp-Stefan

Post Reply
User avatar
David.P
User
Posts: 1501
Joined: Thu Feb 28, 2008 8:16 pm

Multi-language OCR can take a long time

Post by David.P »

Hello Forum and Tracker Support Team,

It is excellent that the OCR feature of PDF-XChange Editor can handle multiple languages simultaneously.

However, I noticed that OCR sometimes takes a long time to recognize text.

When multiple languages are selected such as shown below, for example with the attached document, PDF-XChange Editor takes about 1 minute per page on a fairly powerful PC. Admittedly, the quality of the document is poor, so trying to recognize multiple languages is probably particularly challenging.

image.png
image.png

So is this relatively low OCR speed expected behavior, or is there possible room for improvement?

Thank you
David

Slow Multilanguage OCR.pdf
(755.05 KiB) Downloaded 33 times
David.P
PDF-XChange Pro
User avatar
Dimitar - Tracker Supp
Site Admin
Posts: 1778
Joined: Mon Jan 15, 2018 9:01 am

Re: Multi-language OCR can take a long time  SOLVED

Post by Dimitar - Tracker Supp »

Hello David,

Thank you for your report.

There is an improvement coming up for the OCR tool that will affect its performance.

Currently, the OCR tool is not using the system resources in full, so probably this is the reason for the issue you are facing.

Regards.
User avatar
David.P
User
Posts: 1501
Joined: Thu Feb 28, 2008 8:16 pm

Re: Multi-language OCR can take a long time

Post by David.P »

Thank you Dimitar -- am looking forward to it.

:)
David.P
PDF-XChange Pro
Sasha - Tracker Dev Team
User
Posts: 5522
Joined: Fri Nov 21, 2014 8:27 am
Contact:

Re: Multi-language OCR can take a long time

Post by Sasha - Tracker Dev Team »

Hello David.P,

Yeah - that will be included in the release video that should be out in a couple of hours ;)

Cheers,
Alex
Subscribe at:
https://www.youtube.com/channel/UC-TwAMNi1haxJ1FX3LvB4CQ
User avatar
David.P
User
Posts: 1501
Joined: Thu Feb 28, 2008 8:16 pm

Re: Multi-language OCR can take a long time

Post by David.P »

I'm blown away by the performance of the revamped OCR feature!

I just did a OCR test with 27 pages ("Searchable Image", English, accurracy "Automatic"), and got the following results:

  • PDF-XChange Editor v.9 Build 351: 70 seconds
  • Adobe Acrobat Pro v.11: 70 seconds
  • Adobe Acrobat Pro DC v.2019.008: 60 seconds
  • PDF-XChange Editor v.9 Build 354: 11 seconds

ImageImageImage
David.P
PDF-XChange Pro
User avatar
Paul - Tracker Supp
Site Admin
Posts: 6813
Joined: Wed Mar 25, 2009 10:37 pm
Location: Chemainus, Canada
Contact:

Re: Multi-language OCR can take a long time

Post by Paul - Tracker Supp »

Thanks for posting this David,

I am really pleased to see this translate into real world result. It is one thing to perform well in tests in a controlled environment but the real proof of the pudding is you.

I am impressed with the work they did also, and would offer a big pat on the back to the team for pulling this off!
Best regards

Paul O'Rorke
Tracker Support North America
http://www.tracker-software.com
Post Reply