Skip to content

How to edit a scanned PDF (without retyping the page)

Updated

You open a scanned PDF, click on a word to fix it, and nothing happens. Or the editor lets you draw a text box, but the old words are still there underneath. That is not your editor being broken. A scan is a different kind of PDF, and it needs a different kind of editing.

Why a scanned PDF will not let you edit it

A PDF made from Word, a web page or an invoicing system stores its text as text: letters, a font, and a position on the page. An editor can select those letters and change them.

A scanner does not know about letters. It takes a photograph of the paper and wraps that photograph in a PDF. What looks like a sentence is a pattern of grey and black pixels, exactly like a photo of a sign. There is nothing to select, so there is nothing to edit.

Some scanners and phone apps run OCR (optical character recognition) as they scan. OCR reads the letters off the picture and hides an invisible layer of text behind it, so you can search and copy. That hidden layer still does not change what you see: edit it, and the picture on top stays the same.

The usual workarounds, and what they cost

Convert it to Word. Word, Google Docs and most PDF-to-Word converters run OCR and rebuild the page as a document. For a plain letter this can work. For anything with a layout (a certificate, a form, a letterhead, a table), the result rarely looks like the original. Fonts are guessed, spacing drifts, logos move, and you end up rebuilding the page by hand.

Cover it and type on top. Draw a white box over the old words and put a text box over it. This is quick, but it shows: the box is whiter than the paper around it, the new font does not match, and on patterned or tinted paper it looks like a sticker.

Retype the whole page. It gives a clean result, but it takes the longest and nothing about it matches the original.

Editing the text in place

The approach that looks right does three things for the line you change:

  1. Reads the line with OCR, so it knows which letters are there and how big they are.
  2. Lifts the old letters out of the picture and rebuilds the paper behind them from the paper around them. That includes the tint, the grain and any printed pattern.
  3. Sets the new words in a matching typeface, at the same size, on the same baseline, in the colour of the ink around it.

That is what PencoPDF's scanned PDF editor does. You click a line, retype it, and the change sits in the scan as if it had been printed that way.

Step by step

  1. Open Edit a scanned PDF and choose your file. The words on the page are read as it opens.
  2. Click the line you want to change. It is picked out so you can see exactly what will be replaced.
  3. Type the new wording and click away. The old letters are removed and the background rebuilt behind the new ones.
  4. Check the page, fix anything else, and download. There is no watermark and no account to make.

Getting a good result

What about the original words?

On a scan, the page is a picture. When a line is edited, the rebuilt strip is laid over the scan, so a determined person who pulls the original image out of the file could still find the old words underneath. For fixing a typo or updating a date that does not matter. For removing something confidential it does: use redaction, which blanks the pixels themselves. Whiteout vs redaction explains why.

Only use it on documents you are entitled to change

Changing the wording on your own letter, a form you filled in, or a document you issued is ordinary editing. Changing a certificate, a contract or an official record that someone else issued, to pass it off as theirs, is forgery. The tool does not know the difference; you do.

More guides

Every guide