How to get text out of a scanned PDF

You try to select a line in a PDF and nothing highlights. Or you press search and it finds nothing. That PDF is probably a scan. Each page is a photo of a page, so the computer sees a picture, not words.

How to tell if your PDF is a scan

Open the PDF and try to select a sentence with your finger or mouse. If you can highlight single words, the text is real. If a whole page or a big block highlights at once, or nothing highlights, it is a scan.

Real text? Use PDF to text. It is faster and more accurate. Scan? Read on.

Steps

1. Open OCR scanned PDF.
2. Choose your PDF (up to 25 MB).
3. Pick the language: English, Hindi, or both mixed.
4. Press Read scanned pages.
5. Wait. Each page is read one after another, and the page shows which one it is on.
6. Copy the text or download it as a .txt file. Each page starts with a heading like "Page 1".

Your PDF is read on your own device. It is not uploaded.

More than 20 pages?

The tool reads the first 20 pages in one go. For a longer document, cut it into parts with Split PDF (for example pages 1-20, then 21-40) and read each part.

Why it takes a while

Reading pages is slow work, and it runs on your own phone or laptop. A 20-page document can take a few minutes on an older phone. Keep the screen on and stay on the page until it finishes.

How accurate is it?

It depends on the scan. A clean, straight, black-on-white scan gives very good results. A faded photocopy, a crooked scan, stamps over the text, or handwriting gives worse ones. Check important numbers against the original.

If you are scanning the pages yourself, a few choices help: scan upright, use at least 200 dpi, and pick black and white or greyscale rather than colour.

What next?

• Need a Word file? Paste the text into Text to Word and PDF.
• Need a short version? Paste it into Summarise PDF.
• Need tables in Excel? Scanned tables are hard. The PDF to Excel tool needs real text, so for a scan, retype the figures from the OCR result.