Skip to content
Documentation
View as Markdown

Text recognition (OCR) with Adobe Acrobat Pro

Whether an uploaded PDF file contains text decides whether Librario’s full-text search can find it. If you scanned a publication yourself, or Librario shows the message “Recognised text is short. Please check the file.” on the file, run text recognition (OCR) on the PDF file.

Adobe Acrobat Pro is a program that many offices already have installed and that can recognise text. Don’t confuse it with the free Adobe Acrobat Reader.

Screenshots show Adobe Acrobat XI

The screenshots in this guide come from the German version of Adobe Acrobat XI Pro. Newer versions of Adobe Acrobat Pro have a similar layout; the names of the functions may differ slightly.

The example is a table of contents that Librario downloaded automatically during an import by identifier.

A scanned table of contents in Adobe Acrobat. The text is not yet searchable.
The example PDF file
Screenshot of the German interface.

Step 1: Start text recognition

The Adobe Acrobat toolbar sits on the right-hand side. It holds the functions you use to edit PDF files. Click the “Text Recognition” tab and choose “In This File”.

The Adobe Acrobat toolbar with the "Text Recognition" tab open and the "In This File" option.
Choose Text Recognition: In This File
Screenshot of the German interface.

Step 2: Configure text recognition

Next, set up text recognition:

  • The language of the document: Adobe Acrobat Pro uses a different vocabulary for each language, which helps it recognise the right characters and words.
  • The PDF output style “ClearScan”: according to Adobe, the best option for PDF files that are as small and as little pixelated as possible.
  • Downsample to: set it to the highest possible value.

Close both dialog boxes with “OK”.

The "Recognize Text" dialog box in Adobe Acrobat with the language setting and the PDF output style "ClearScan".
Click "Edit" and choose ClearScan
Screenshot of the German interface.
The progress bar of a running text recognition in Adobe Acrobat.
Text recognition is running and takes a while
Screenshot of the German interface.
The PDF file after text recognition: a line of text is selected, so the file is searchable.
Afterwards you can select lines of text and search the PDF file
Screenshot of the German interface.

Step 3: Further optimisation

Text recognition is really all you need. Don’t forget to save the new PDF file.

You can optimise the PDF file further, though. The aim is to make the file as small as possible: the smaller the file, the faster users can download and open it from Librario later.

Use the “Optimize Scanned Documents” action in the Action Wizard for this. In the toolbar, click “Action Wizard” and then “Optimize Scanned Documents”.

The Action Wizard in Adobe Acrobat with the "Optimize Scanned Documents" action.
Action Wizard: Optimize Scanned Documents
Screenshot of the German interface.

The wizard runs three steps:

  • Add Document Description
  • Optimize Scanned PDF
  • Save As

Click “Start” to run the wizard.

Step 3a: Add a document description

Enter the title of the publication here, and the authors in the “Author” field. Librario reads this information back later as PDF metadata.

The "Add Document Description" dialog box with fields for title and author.
Add Document Description
Screenshot of the German interface.

Step 3b: Optimise the scanned PDF

This step runs on its own and optimises the internal structure of the PDF file.

The Action Wizard optimising the scanned PDF file automatically.
Screenshot of the German interface.

Step 3c: Save the file

Finally, save the new file. You can overwrite the old file or save the new one under a different name.

The "Save As" dialog box at the end of the Action Wizard.
Screenshot of the German interface.

The result is a much smaller file (46% smaller in the example) that also looks better and that Librario’s full-text search can read.

The original file next to the final result at a zoom of 1,600%: the final result is sharper.
The original file (left) compared with the final result (right) at 1,600% zoom
Screenshot of the German interface.

For general advice on searchable files, such as other OCR programs, scan settings and quality checks, see Optimising files for full-text search.