Image to Text (OCR)

Have a picture with words in it — a photo of a page, a screenshot, a receipt or a sign? Read the text out of it and download it as a plain text file, or as a searchable PDF that keeps the image and makes its words selectable.

Files are never uploaded

How to use it

Add one or more images

Drop your pictures onto the box above, or click to browse. You can add several at once; each one becomes its own text file.

Choose the text layout

Leave it on Automatic for most images. Use "A single line" for a screenshot of one row, or "Sparse text" when the words are scattered across a sign or a label.

Let it read the words

The tool looks at the pixels and recognises letters, numbers and punctuation. The progress bar moves as each image is processed.

Check the result

Open a text file and compare it with the picture. Small or blurred text and fancy fonts are where mistakes happen, so check anything you will rely on.

Download the text files

Save the files individually, or use Download all when you added several images. Tick “Also save a searchable PDF” if you want the picture back with selectable text on top.

Common issues

Blur and glare defeat recognition

A photo taken in a hurry is often slightly out of focus or has a bright reflection across the words. Retake it square-on, in even light, and run it again.

Rotated photos read badly

Text that is sideways or upside down is misread or missed. Straighten the image before adding it — the recogniser expects the text roughly level.

Numbers and similar-looking letters get swapped

Characters such as 0 and O, 1 and l, 5 and S are easily confused. Always proofread figures you intend to use.

Plain text keeps no layout at all

The .txt output carries words only — no fonts, colours, columns or images from the picture. If you want the picture preserved, tick the searchable PDF option, which keeps the image and hides the text behind it.

Frequently asked questions

What images work best?
Sharp, well-lit, upright images with clear contrast — a photo of a printed page, a screenshot, or a flatbed scan. Straighten the picture and fill the frame with the text for the best result.
What is the searchable PDF option for?
It returns your picture as a PDF with an invisible text layer behind it, so the image looks untouched but the words can be selected, copied and searched. It is the equivalent of making a paper scan searchable.
Which languages are supported?
English only for now. The English language data is bundled with the tool, so recognition runs offline and nothing is fetched from a third party.
Are my images uploaded?
No. Every image is read inside your browser. Nothing is sent to a server, and neither the recognised text nor any PDF is sent anywhere.
Can it read handwriting?
Poorly, and it cannot be relied on. This kind of OCR is built for printed, machine-set text. Clear block capitals on a form sometimes work; joined-up writing generally does not.
Why is a small logo or a stylised heading missed?
Decorative fonts and letters that are part of a graphic are hard to recognise. Try the "Sparse text" layout, or crop the image so the words you need fill more of the frame.
Can I process several images at once?
Yes. Add as many as you like and each one is recognised in turn. Every image produces its own text file, and its own searchable PDF if you turned that option on.
Does the text keep the arrangement of the words?
No. Text is returned in reading order, and the same applies to the text layer behind a searchable PDF. Columns, tables and side-by-side text are flattened into a single stream.
What if no text is found?
The tool tells you so. Check the image really does contain readable words, that it is not rotated, and that the text is not too small, then try again.

Related tools

Send us feedback

Something broken, or unclear? Tell us and we will look at it.

About this page. Written and maintained by the AI-Mind team. Last reviewed 11 October 2026. This tool runs entirely in your browser. Found an error or have a suggestion? Use the feedback form above or email [email protected].