Scanned PDF to Text: How to Get Editable Text from a Scan

CorrectNow Editorial Team · 5 min read

You open a PDF, try to copy a paragraph, and nothing happens, or the whole page highlights as one block. That usually means the PDF is a scan: a set of page images saved inside a PDF file. It looks like a document but contains no actual text.

To edit, search or reuse the words, you need OCR. This guide explains how to recognise a scanned PDF, how to prepare it, how to extract the text and what to fix afterwards.

Text PDFs and scanned PDFs are different

A PDF created from a word processor or exported from a website normally contains real text. You can select single words, search with Ctrl+F and copy sentences.

A scanned PDF is made from images, from a scanner, a photocopier or a phone scanning app. Every page is a picture. Search finds nothing, and copying either does nothing or copies an image. Some PDFs are mixed, with a few typed pages and a few scanned ones, so test more than one page.

There is also a third kind: a scanned PDF that already has an invisible OCR text layer. If you can select text on a page that is clearly a scan, someone has run OCR on it before. That text may still contain errors, so check it.

Prepare the PDF for better OCR

OCR quality depends heavily on scan quality. If you are scanning the pages yourself, a few choices help:

  • Scan at a resolution high enough for small print to look sharp; 300 dpi is a common recommendation for ordinary text.
  • Keep pages straight on the glass so lines are not tilted.
  • Use black and white or greyscale for plain text documents, and colour only if colour carries meaning.
  • Remove blank pages and pages you do not need.
  • Split very large files into smaller parts if a tool has a size limit.

Extract the text with OCR

Upload the PDF to an OCR tool, let it process the pages and review the extracted text. On CorrectNow's OCR page you can upload PDFs up to 8 MB, and the text keeps its original language and script, which matters for documents that are not in English.

If a file is too large, split it by chapter or by a range of pages and process each part. This also makes checking easier, because you review one manageable section at a time.

Fix layout problems after extraction

OCR reads words well but can lose the structure of the page. Expect to tidy up a few things:

  • Two-column pages may come out with lines from both columns mixed together. Check that paragraphs read in the right order.
  • Headers, footers and page numbers repeat on every page and often land in the middle of paragraphs. Delete them.
  • Words broken with a hyphen at the end of a line may stay split. Join them back together.
  • Tables may turn into loose lines of text. Rebuild important tables by hand.
  • Footnotes can end up in the body text. Move them to the end or remove them.
  • Incorrect: The committee approved the infor- mation policy. – Correct: The committee approved the information policy. (A line-end hyphen kept by OCR splits one word into two.)

Check the words, not just the layout

Once the structure is right, read for recognition errors. Look closely at numbers, names, dates and anything in small print. Similar shapes are the usual trouble: 0 and O, 1 and l, rn and m, cl and d.

Running the cleaned text through a grammar checker helps catch missing letters and broken words that are hard to see when you are focused on layout. For contracts, financial records or official papers, compare the final text with the scan line by line.

Keep the scan alongside the text

Save the original scanned PDF even after you have editable text. It is your reference if a question comes up later about what the document actually said, and it preserves signatures, stamps and handwriting that OCR does not capture.

Give the text file a name that matches the scan, so the two stay together. If you share the editable version with others, mention that it was produced with OCR, so readers know to check the original for anything critical.

Frequently asked questions

How do I know if my PDF is scanned?

Try to select a single word or search for a word you can see on the page. If neither works, the page is an image and needs OCR.

Can I convert a scanned PDF to Word?

You can extract the text with OCR and paste it into a Word document. Complex layouts, tables and columns usually need some manual formatting afterwards.

Why does my extracted text have random line breaks?

OCR often keeps the line breaks from the printed page. Remove them so sentences flow as normal paragraphs, and watch for hyphenated words at line ends.

Will OCR keep the original language of my PDF?

A good OCR tool keeps the text in its original language and script. Translation, if you need it, should be a separate step after checking the text.

Extract text from a scanned PDF · Proofread the extracted text · Browse CorrectNow tutorials

Related articles