Skip to main content
Participant
October 17, 2016
Question

improve OCR results

  • October 17, 2016
  • 1 reply
  • 1478 views

I often scan in documents to Acrobat Pro, and then run OCR on them. I want to copy and paste excerpts into other documents. However, the results of the OCR are very poor -- for example it does not recognize the "fi" combination or other combinations that typically appear ligatured. Is there any way to improve the results other than manual correction after the cut and paste?

This topic has been closed for replies.

1 reply

Lovekesh Garg
Adobe Employee
Adobe Employee
October 18, 2016

You can update anything recognized by Acrobat after OCR using suspect correction.

For Acrobat XI ir older versions

Go to tools > Text Recognition> Find First suspect. Correct the text if it is marked as suspect

For Acrobat DC:

Go to Tools> Enhance scans> Recognize Text> Correct recognize text. Correct the text if it is marked as suspect

Also if a word is not marked as suspect and you still think its not correct, click on the checkbox "Review recognize text" and double click on the word you want to change.

Also please share the Acrobat version you are using and a sample PDF where you are facing this issue, so that we can better look into this issue and resolve it. You can use https://cloud.acrobat.com/send​ to share the file.

Thanks.

Participant
November 29, 2018

I'm having the same issue where I'm pretty sure Acrobat is not recognizing my text correctly, but I can't tell for sure. When I try to "review recognized text" it doesn't allow me to double click on any words because the screen grays out. I took a screen shot and attached the PDF I'm trying to optimize. Shared Files - Acrobat.com

How do I know if a reader will correctly read this document, and how can I change text that Adobe Acrobat did not recognize correctly? I'm using Pro DC.