ES

Run OCR on a PDF

Extract English text from scanned PDFs with OCR and download a transcript. Review recognition errors; the original PDF does not become searchable.

Extract real, selectable text from scanned or image-based PDFs, so you can search, copy, and paste content that used to be locked inside a picture.

What this tool can do

Extract text from scanned PDF pages using English OCR. Fileixa returns recognized text; this tool does not add a searchable text layer to your original PDF. Recognition depends on scan quality, so proofread names, dates and numbers before using the output.

What OCR does

A scanned PDF is just pictures of pages, so you cannot search it or copy its text. Optical character recognition (OCR) reads the letters in those pictures with the Tesseract engine and returns the text.

Accuracy depends mostly on the scan. Clean, straight, 300 dpi black-on-white pages come out very well; skewed phone photos, low resolution, handwriting and decorative fonts come out worse.

Getting better results

Rotate pages upright first, scan at 300 dpi rather than 150, and crop away dark borders. OCR is currently tuned for English text.

How to use this tool

  1. Upload the scanned or image-based PDF.
  2. Wait while OCR reads the text on every page.
  3. Download the result with searchable, selectable text.

Frequently asked questions

Does this work on handwriting?

OCR is most reliable on printed text; handwritten text may be recognized inconsistently depending on legibility.

Do I need this for a normal, text-based PDF?

No — if you can already select text in the PDF, it does not need OCR; this tool is for scanned pages or photographed documents.

Guides for this task

Related tools

Merge PDFSplit PDFCompress PDFRotate PDFView all: PDF Tools