Here’s the transcript:
In this video, we’ll look at how to correct OCR errors. Acrobat does a surprisingly good job of capturing the text, but it’s best to review and correct the errors.
Here’s my document. I’ve already captured the text. If I copy the text and paste it into a Word document, you can see there are a fair number of issues. Let’s fix them.
Correcting the Text
Back in Acrobat, I have the Enhance Scans tools available. Click the Recognize Text command and choose Correct Recognized Text. You’ll see a message open explaining that possible suspects are highlighted.
A suspect is any string of text that Acrobat can’t define with certainty.
Close the message box and let’s start.
You’ll see the first suspect highlighted. The Image of the text and its interpreted content display. Click Accept to confirm the interpretation and move to the next. When you come to a string of text that isn’t correct, type the replacement in the ‘recognized as’ field, and click Accept to continue.
Here’s a quick tip: Rather than clicking Accept each time, just press ENTER to go to the next suspect.
If you want to keep track of your progress, click Review Recognized Text to add an overlay to the page showing the captured content. What if Acrobat missed a word? No problem. As long as you’re seeing the recognized text overlay, just double-click the word that needs correcting, and type in its correct letters.
Here’s my finished captured page, and here’s the Word doc.