How to OCR a Scanned PDF on iPhone

TL;DR

For scanned or image-based PDFs with no real text layer, PDFExtract uses on-device OCR to read the pages into text — so even a photographed document becomes searchable, editable content without uploading anything.

Steps

Import the scanned PDF into PDFExtract.
Run text extraction — PDFExtract uses OCR where there’s no text layer.
Set the OCR language(s) to match the document.
Preview the recognized text as Markdown.
Share or export the result.

Why it works this way

A scanned PDF looks like text but is really a picture — copy-paste gets you nothing. On-device OCR reads the characters out of the image so the page becomes real, searchable text. Keeping OCR local matters here: scanned documents are often exactly the sensitive kind (IDs, contracts, records) you don’t want on a server.

Tips & edge cases

Set the right OCR language before running for better accuracy.
Clean, straight scans OCR best — skewed or low-contrast pages are harder.
Everything is on-device, so scanned records stay private.

FAQ

What if my PDF is just scanned images? PDFExtract falls back to on-device OCR to read text from scanned or image-based pages.

Does OCR happen in the cloud? No. OCR runs on your device, so scanned documents stay private.

Can I improve OCR accuracy? Yes. Adjust OCR languages, language detection, and correction settings to match your document.