Image OCR

Image OCR (optical character recognition) reads the text in an image and turns it into text you can copy, edit and search. It works on screenshots, scanned pages and photos of documents, signs and slides, in JPG, PNG or WebP. Recognition runs in your browser with the open-source Tesseract engine, so the image is never uploaded to a server.

How to extract text from an image

1. Choose an image or drag it onto the page.
2. Click the extract button. The first run downloads the recognition engine and English language data, which takes a few seconds.
3. Check the recognised text in the result box, correct any mistakes, and click Copy.

Tips for the most accurate text

• Use a sharp, well-lit image. Blurry photos, shadows and glare cause most recognition errors.
• Keep the text straight. Rotate sideways photos first with Image Rotate, and photograph pages from directly above.
• Crop to the text. Removing borders, pictures and busy backgrounds with the Image Cropper helps the engine find the lines.
• Bigger text is better. Small print in a low-resolution screenshot is hard to read; zoom in before taking the screenshot.
• Printed text works best. Handwriting is recognised poorly by any OCR engine.

When it comes in handy

• Copying text from a screenshot, a slide or an image you received.
• Turning a photo of a receipt, letter or page into editable text.
• Getting quotes out of book pages and printed articles.
• Making text in images searchable and accessible.

Working with PDFs? PDF to Text extracts the text of PDFs that already contain a text layer. For a scanned PDF, save the pages as images with PDF to PNG first, then run OCR on them here.

Frequently asked questions