imissfiles56 tools
  1. Home
  2. OCR tools
  3. OCR PDF

OCR a PDF

Turn a scan into a searchable PDF, or get the text out as plain .txt. Recognition runs on this device.

OCR a scanned PDF

Never uploaded

Your file is processed in this tab and never sent anywhere.

The ads come from Google. Your files stay on your device.

How to OCR a scanned PDF

  1. Drop a scanned PDF. The kind where you cannot select any text.
  2. Let it recognise each page. Pages are rendered and read one at a time on this device.
  3. Download. A searchable PDF that looks identical to the original, or the plain text as .txt.

About OCR PDF

There are two completely different kinds of PDF, and knowing which one you have saves a lot of time. A PDF exported from a word processor carries an invisible text layer, which is why you can select and search it. A PDF produced by a scanner or a phone is a photograph of paper wrapped in a PDF container: there is no text in it at all, only pixels arranged to look like text.

The test takes two seconds: open the file and try to select a sentence. If your cursor highlights words, you have a text layer and <a href="/pdf-to-text">PDF to Text</a> will do a faster and more accurate job than this tool, because it reads exactly what is stored rather than guessing from shapes. If nothing highlights, you have a scan, and recognition is the only option.

This tool renders each page with PDF.js at a resolution tuned for recognition, then runs Tesseract over the resulting image. Both run in your browser tab. It is slower than text extraction by a wide margin (expect a few seconds per page on a laptop and longer on a phone) because recognising letter shapes is real computation in a way that reading stored characters is not.

There are two outputs, and they answer different needs. The searchable PDF is the scan exactly as it was (every pixel of every page) with the recognised words laid invisibly on top at their own positions. It looks identical, prints identically, and you can now search it, select text from it, and copy from it. That is what people usually mean by "OCR PDF to PDF", and it is the right choice for archiving, for anything you need to find again, and for documents someone else will search. The plain text output is for when you want the words themselves: to quote, to paste, to feed into something else. Columns are read in the order the engine finds them, which for a two-column page is sometimes wrong, and tables come out as lines rather than grids.

OCR PDF FAQ

Is there a page limit?

Yes: 200 pages for the searchable PDF, because every rendered page is held in memory until the file is saved. For a longer scan, split it into parts of 200 or fewer with Split PDF, run each, then merge the results.

How do I know if my PDF needs OCR?

Try selecting a sentence in your PDF viewer. If text highlights, use PDF to Text instead: it is faster and exact. If nothing highlights, the file is a scan and needs OCR.

Why is this so much slower than PDF to Text?

Text extraction reads characters already stored in the file. OCR renders each page as an image and works out what the letter shapes are, which is far more computation.

Does this create a searchable PDF?

Yes. The searchable PDF output keeps every page exactly as scanned and adds the recognised words as an invisible text layer at their positions, so the document looks the same but search, select, and copy now work.

What does "OCR PDF to PDF" mean?

Turning a scan into a searchable PDF: same appearance, but with a hidden text layer added so it can be searched and selected. That is the searchable PDF output here.

Will tables come out correctly?

Not as tables. You get the text in reading order. For tabular data from a scan, Image to Excel does a better job on a single page.

How accurate is it?

Good on clean 300 DPI scans of printed text. Worse on low-resolution scans, skewed pages, and anything handwritten.

Is my document uploaded?

No. Page rendering uses PDF.js and recognition uses Tesseract as WebAssembly, both in this browser tab.

Related tools

Related reading