Image Tools

OCR Extractor – Extract Text from Images (Free & Private)

Turn photos, screenshots and scans into text you can copy, edit and download.

Add an image, choose the language and copy the text. This free OCR extractor works entirely in your browser, so your images are never uploaded.

Your images never leave your device

Drop images here, or choose them.PNG, JPG, WebP, BMP or GIF. Up to 10 images, 15 MB each.

Your image stays on your device. Nothing is uploaded.

The first run loads the text engine (about 7 MB), then it works offline.
Images and text
Your images and the text found in them will appear here...

What is OCR, and why use it?

OCR, short for optical character recognition, reads the letters in a picture and turns them into text you can select, search and edit. It saves retyping when you have a screenshot, a photo of a page, a scanned document or an image of a receipt. This OCR extractor works online, with nothing to install, and it runs in your browser so your images stay on your device.

How to Extract Text from an Image

It takes three steps, and there is nothing to install or sign up for.

1. Add your images

Drop one or more images onto the upload area, or choose them from your device. Photos, screenshots and scans all work.

2. Extract the text

Pick the language of the text and click Extract Text. The tool reads each image and shows the text next to a preview.

3. Copy or download

Fix any mistakes right in the text box, then copy the text or download it as a .txt file.

How OCR Works

An OCR engine looks at the shapes in an image and works out which characters they are. This tool uses the open-source Tesseract engine, compiled to WebAssembly so that it can run quickly in your browser.

Prepare the image

The image is converted to greyscale. Very small images are enlarged and very large ones reduced, so letters are the right size to read.

Find the lines of text

The engine finds where the text is on the page, and splits it into lines and words, even when the layout has several blocks.

Recognise the letters

A trained model reads each line and picks the most likely characters. The confidence score shows how sure it is.

Supported Image Formats

PNG and JPG

The best choices. Screenshots work well as PNG, and photos of documents as JPG. Transparent areas are read as white paper.

WebP and BMP

Both are read the same way as PNG and JPG. Use whichever your image was saved in.

GIF

A GIF is read from its first frame, so animated GIFs give the text of the first picture only.

Your Images Stay on Your Device

Everything on this page runs in your browser. Your images and text are never uploaded, logged or stored, which makes it safe for receipts, ID cards, contracts and anything else you'd rather keep to yourself. It's also fast, because there's no waiting on a server, and it keeps working even if your connection drops after the page has loaded.

Works on Mobile and Desktop

The OCR extractor works in any modern browser on phones, tablets and computers, with nothing to install. On a large screen each image sits beside its text. On a phone they stack, and every button is sized for touch. You can pick photos straight from your camera roll, or take a new photo of a document and read it right away.

Frequently Asked Questions

How do I extract text from an image?
Add your image, choose the language of the text, and click Extract Text. When it finishes, the text appears next to the image, ready to copy or download as a .txt file.
Is my image uploaded anywhere?
No. The OCR engine runs in your browser, on your own device, and it is loaded from this site. Your images and the text found in them are never uploaded, logged or stored, so it is safe to use with receipts, ID cards and private documents.
How accurate is it?
On clear, printed text it is usually very accurate. Accuracy drops with blur, glare, small or stylised fonts, handwriting and busy backgrounds. The confidence score under each result shows how sure the engine is, and you can fix mistakes in the text box before you copy it.
Which languages are supported?
English, Spanish, French, German, Italian and Portuguese. Choose the language of the text in your image before you click Extract Text, because the engine reads best when it knows what to expect.
Can it read handwriting?
Not reliably. The engine is built for printed text, so neat block capitals sometimes work, but cursive handwriting usually doesn't.
Why does the first run take longer?
The first time you extract text, your browser loads the OCR engine and the language data, about 7 MB for English. After that it is quicker, and the engine runs on your device with no upload. Large images and slower phones take a little longer to read.
How can I get better results?
Use a sharp, well-lit image with dark text on a plain background. Hold the camera straight on to the page, crop away the edges, and avoid shadows and glare. Larger text reads better than very small text.