Extract text from images: photos, screenshots and scans, with free OCR that runs fully in your browser.
Your image
Pick a photo, screenshot or scan. Text is read locally.
Drop an image here
or click to browse. PNG, JPG, WebP and more.
The OCR engine loads from a CDN on first use, then recognition runs on your device.
Recognition
0
Upload an image to begin.
Advertisement
How OCR works here
OCR (optical character recognition) looks at the pixels of your image and identifies letters, numbers and punctuation. This page uses Tesseract.js, a proven open-source OCR engine. The engine library downloads from a CDN the first time you press Extract text; after that, the actual text recognition runs entirely on your device, so the image content never reaches a server.
Getting the best results
Sharp images win: clear, well-lit photos recognize far better than blurry ones.
Straight text helps: avoid heavy angles, shadows and decorative fonts.
Screenshots are ideal: crisp screen text is what OCR handles best.
English text: this page recognizes English. Mixed or handwritten text may be less accurate.
Recognition can take 10 to 30 seconds on large images. The progress bar shows what is happening.
Upload or drop an image, then press Extract text. The OCR engine reads the image and the recognized text appears in the output box, ready to copy.
Is the OCR really free?
Yes. Tesseract.js is open-source and free. There are no limits, accounts or watermarks on this page.
Are my images uploaded anywhere?
No. After the engine library loads from a CDN, recognition runs 100% on your device. The image content never leaves your browser.
What if the text recognition is wrong?
OCR works best on sharp, well-lit, straight-on text like screenshots. Blurry photos, heavy angles, decorative fonts and handwriting reduce accuracy. Try cropping to just the text area.
Which image formats are supported?
Any format your browser can display: PNG, JPG, WebP, GIF and BMP. Screenshots usually give the cleanest results.