TEXT GOES WONKY when go to edit my scanned document.

[Carina Ball]

My scanned Document is a master doc we repeatedly open and edit then do a "SAVE AS". Problem is it make the document illegible and unuseable as it converts elements, handwritten and other- to illegible characters, and lines into dashes and so forth. I need to maintian the integrity of the document.

Can someone advise how to do this?

[Document Geek]

Ideally, you'd work on the native file, as opposed to a scanned original, edited in Acrobat. That's honestly a recipe for trouble.

What kind of edits are you trying to make?

[Bevi Chagnon - PubCom.com]

Scanned documents are pictures of the text.

It's possible to run Acrobat's "Recognize Text" utility, which is OCR (optical character recognition) that tries to interpret the graphics into editable letters and words, but it's dicey and full of problems like those you're experiencing. OCR also doesn't recognize the original font that was used, so you'll end up with illegible characters, especially if it's edited.

And if you edit the live text, the original graphic of the text is still visible so it just gets superimposed above the graphic, making a hot mess.

Two solutions:

1. Find the original native file (such as MS Word), make the edits there, and re-export it to PDF.
2. After you run OCR on the scan, export it to a Word file where you can work on the content.  File / Export To / MS Word.docx usually does a decent job of capturing the live text, graphics, and appearance of the original, but you probably will need to make some adjustments.

[Eric Dumas]

It makes sense if the image of text is re-compressed at every save. I would check the compression settings because a jpeg of a jpeg of a jpeg of a jpeg of a jpeg of a jpeg... will end up completely impossible to read, even after OCR.