OCR PDF: make a scanned PDF searchable
Your scanned PDF, with text you can search and copy.
Add to favourites
Create a free account to save your favourite tools and have them on any device.
Advertisement
A scanned PDF, or one made from phone photos, is just a picture of each page: you can’t search for a word or select and copy the text. OCR (optical character recognition) reads those pages and adds the text to the same PDF, invisible and right under each word, without changing anything you see. Free, with no sign-up and no watermarks.
How to make a scanned PDF searchable
- Select the scanned PDF (up to 30 pages).
- Choose the language the document is written in; English by default.
- Click “Make searchable” and wait: each page takes a few seconds.
- Download the PDF and try searching for a word with Ctrl+F (or Cmd+F on a Mac).
Advertisement
What you get
The same PDF, page by page, with an invisible text layer over every scanned page. What you see doesn’t change: the page images aren’t compressed again, so the quality is the same and the file grows by only a few KB per page. What changes is what you can do with it:
- Search for any word with your PDF reader’s search box.
- Select and copy the text to paste it into an email or another document.
- Let Windows or Mac search, Google Drive or your company’s document system find it.
If the PDF mixes scanned pages and pages with text (say, a report with scanned appendices), we only recognize the scanned ones; the others stay as they were. If every page already has text, we tell you instead of processing it.
For the text to be recognized well
- Choose the document’s language: it gets accents and the letters of each alphabet right. If it mixes two languages, choose the main one.
- Sharp, straight scans: 300 dpi is ideal, and 200 dpi usually works well. Blurry photos, shadows or very skewed pages cause errors.
- Pages upside down or sideways: straighten them first with Rotate PDF, because upside-down text isn’t recognized.
- Handwriting isn’t recognized well: OCR is designed for printed text.
Recognition is never perfect: on a poor scan it may mix up a letter here and there. Check the copied text before using it for anything important.
Advertisement
If you want to edit the text
OCR makes the PDF searchable, not editable: the text is still part of the page image. To change the text, use PDF to Word OCR, which gives it to you in paragraphs you can edit. And if you need to archive the document or file it with a public body, after OCR you can convert it to PDF/A without losing the recognized text.
Limitations
- One PDF at a time, up to 50 MB and 30 pages.
- Password-protected PDFs can’t be processed: remove the protection first with Unlock PDF.
- Very large pages (plans, posters) are read at a lower resolution so they don’t go over 40 megapixels per page.
Frequently asked questions
What is a searchable PDF?
A PDF that has text as well as the picture of each page: you can search for words, select them and copy them. PDFs created from Word already are; scanned ones aren’t until OCR recognizes their text and adds it to the PDF.
How do I know if my PDF needs OCR?
Open it and search for a word you can see on the page with Ctrl+F, or try to select a line. If it isn’t found or you can’t select the text, the PDF is an image and needs OCR. If it already has text, this tool will tell you.
Can you see the added text, or does the PDF look different?
No. The recognized text is invisible and placed over each word of the image, so the PDF looks and prints exactly as before. You only notice it when you search, select or copy.
Does it work with handwritten documents?
Not well. Recognition is designed for printed or typed text; with handwriting it recognizes little or nothing. On printed forms filled in by hand, it recognizes the printed part but not the handwriting.
Can I OCR a PDF in several languages?
Text is recognized in one language at a time, so choose the one the document is mostly written in. Words in other languages that use the same alphabet are usually recognized too, though an accent or a letter specific to that language may come out wrong.
Advertisement
Convert your PDF files too
Advertisement
Advertisement