Copy text in pdf gives me gibberish. Is there a way to OCR to correct?

[alanterra]

I have a few documents that are complete gibberish when I select text and copy. If I open them in Acrobat Pro, select text, Copy, and "Show Clipboard" in Finder, I see a bunch of "skull characters", and if I open them in Preview and do the same, I see strings of dots. The text in the clipboard cannot be pasted intelligibly into any other program, and I cannot search the document.

Some of these documents were downloaded from commercial sites. One of them came from (I think) OCR'ing a scanned document using ClearScan. An example of such a document is at https://public.me.com/ix/alanterra/Reynoso%202006%20p%201.pdf?disposition=download+1317001233647 (it's small, 55K).

It seems to me that one way to do this would be to convert the document into a "scanned" pdf, and then OCR it. But the only way I can figure out how to do this is to image each page separately in Photoshop, and then assemble the pages into a new document.

There must be a way to deal with this problem.

Any thoughts?

A

PS--If you look at the document linked to above, you will note that the text in the footer is coherent, but not the text in the body of the document.

[Luke_Jennings · Accepted solution]

You will be able to use OCR in Acrobat after you convert the type to outlines. You will need to add some transparency, then use the flattener preview to outline your type. Here are the steps (for Acrobat 9):

1. Document> Watermark> Add (add a text watermark, hit the space bar once).

2. Advanced> Print Production> Flattener Preview> Convert all text to outlines (checkbox on). Save.

3. Document> OCR text recognition> recognize text using OCR. Select all text with the type tool, copy.

This method is not perfect, you will need to check the copy for errors.

[alanterra]

Luke,

You are a genius

Thanks!

[MetroCarla]

This worked for me and saved me tons of time! Thank you very much, Luke!

[Sara5FD6]

This is great, thank you! Even in 2021 it was the most helpful google result. It's not perfect indeed but it saves a lot of typing time, thanks!

[Wardo31988500axty]

how did you find this out I have been trying for 7 hour straight embeding fonts and stuff on a 700 page file and you hit with 4 random steps that works like magic I litrally did your method as a joke expecting that it wont work. You need a medal  

[Mell17]

Try selecting the text, right-clicking it, and choosing "copy with formatting". This works for me.

[ProlificUser]

Thanks for the solution Mel. 

[Pedro5C2D]

Right-clicking only gives me the options: "Copy", "Highlight Text", "Add Note to Replace Text" and "Add Note to Text", the "copy with formatting" option doesn't appear for me

[Mell17]

There are some suggestions about enabling "Copy with formatting" here:

https://community.adobe.com/t5/acrobat/quot-copy-with-formatting-quot-option-doesn-t-appear/td-p/970...

Hopefully that will work for you.

[defaultynmriukqx0d2]

Thanks! This was helpful.

However, you can't even search the text that you copied from the document.

I think this is because the Acrobat search string is plain text (the gibberish search characters). For example, I used "Copy with formatting" to copy a text string, then searched using that text string (quotation marks properly removed), and got no results. 

FWIW, the document language is Japanese.

The unformatted text string is: 隨ｬ6谺｡謾ｹ險ら沿

Copy with formatting: 第6次改訂版

Result of search: "Adobe Acrobat has finished searching the document. No matches were found."

Result of result: /facepalm

Don't suppose there's a genius out there with a workaround?

[Vladimir5C90]

Hi All,

After trying a bunch of different things that didn't work on the free version of Adobe Reader I ended up installing Foxit Reader (free version) and it worked out immediately.