How to Extract Text from an Image or Screenshot (Free OCR)

You've got text trapped in an image — a screenshot you can't select, a photographed document, a printed page you need to edit. OCR (Optical Character Recognition) pulls that text out so you can copy, edit and search it.

Do it free, without uploading

The image to text converter runs OCR entirely in your browser using Tesseract — your image never leaves your device. That matters when the image is a contract, an ID, or anything you'd rather not upload to a stranger's server. Upload the image, click Extract, copy the result.

What it reads well — and what it struggles with

Clear printed or typed text extracts very accurately: screenshots, book pages, printed notices, receipts. It struggles with handwriting (especially cursive), stylized fonts, low-resolution images, and text at angles. Numbers and standard symbols come through fine.

Getting the best accuracy

Four things make a big difference: resolution (higher is better — text under ~20 pixels tall fails), straightness (rotate skewed images first with the image rotator), cropping (trim to just the text region using the cropper), and lighting (even light, no shadows across the text). Black text on a white background is the ideal case.

Common uses

Copying text from a screenshot, digitizing printed notes for editing, grabbing an error message photographed from another screen, extracting figures from a scanned receipt, or pulling a quote from a book page. First use downloads the OCR engine (~3 MB), then it's cached for instant runs.

After extraction

Clean up the text with the remove extra spaces tool, count it with the word counter, or convert its case with the case converter — all in the same browser session.