N
NJHTools

Image to Text OCR Online

Copy text from any image or screenshot for free: paste a screenshot with Ctrl+V or drop a JPG or PNG, and OCR turns it into editable text in English, Korean, Japanese, Chinese and 6 more languages. Recognition runs in your browser, so your image is never uploaded.

Pick the language in the image. Its data downloads once, then it's cached.

Text layout

Joining removes mid-sentence line breaks. Text you've edited is kept as is.

Preprocessing

Grayscale, extra contrast and upscaling of small images. Applies to the next scan.

Drop images here, click to choose, or press Ctrl+V to paste a screenshot

PNG, JPG, WebP, BMP or GIF. Several images at once is fine. Read on your device, never uploaded.

The OCR engine and language data are downloaded once from a CDN and cached; your image is never uploaded.

How to extract text from an image

  1. Choose the language of the text in your image, for example English, Korean, or English + Korean.
  2. Press Ctrl+V (Cmd+V on Mac) to paste a screenshot, or drag in JPG, PNG, WebP, BMP or GIF files.
  3. Wait while the image is read. The first run downloads the OCR engine and language data, so it takes a little longer.
  4. Edit the text if needed, then click Copy text or Download .txt.

Tips for better OCR accuracy

  • Resolution: text should be clearly readable when you zoom in. Small or blurry screenshots read worse; capture at 100% zoom or larger.
  • Contrast: dark text on a light, even background works best. For photos of paper, turn on "Improve accuracy for photos".
  • Straight text: photograph pages straight on, avoid tilt, shadows and glare, and crop away everything that isn't text with the Image Cropper.
  • Right language: picking the correct language matters most. Use English + Korean for mixed text.
  • Scanned PDFs: OCR reads images, not PDF files. Convert the pages to PNG at 300 DPI with PDF to JPG, then drop the images here.

Supported languages

This online OCR reads printed text in English, Korean, Japanese, Chinese (Simplified and Traditional), Spanish, French, German, Portuguese and Russian, plus an English + Korean mode for mixed documents. Korean users can use it as a free 이미지 텍스트 추출 tool for screenshots, scanned documents and photos of printed pages. Handwriting is only partly supported: neat block letters may work, but cursive and messy notes usually don't.

Privacy: how this image to text converter works

Text recognition runs in your browser with Tesseract, an open-source OCR engine, compiled to WebAssembly (tesseract.js). The OCR engine and language data are downloaded once from a CDN (jsDelivr) and cached by your browser; your image is never uploaded. You can check in DevTools → Network: the engine and language files load, but no request carries your image. Screenshots of chats, receipts, IDs or work documents stay on your device, and there are no accounts, limits or watermarks.

Related Tools

Image to PDFHEIC to JPGImage CompressorImage CropperPDF to JPGWord Counter

Frequently Asked Questions

Can I copy text from a screenshot?↓

Yes. Take a screenshot (Win+Shift+S on Windows, Cmd+Ctrl+Shift+4 on Mac), then press Ctrl+V or Cmd+V on this page. The text is extracted in a few seconds and you can copy it or download it as a .txt file.

Can I convert a JPG or a PDF to text?↓

JPG to text works directly: drop JPG, PNG, WebP, BMP or GIF files and the text is extracted in your browser. PDFs aren't read directly. Convert the pages to PNG at 300 DPI with the free PDF to JPG tool first, then drop those images here.

Does it work with Korean?↓

Yes. Choose Korean, or English + Korean for mixed text, before adding the image. Korean OCR works best on clear printed text such as screenshots, documents and signs.

Is this image to text converter free?↓

Yes, it is completely free with no signup, no daily limit and no watermark. You can process as many images as you like.

Is my image uploaded to a server?↓

No. OCR runs entirely in your browser with tesseract.js. The first time you use it, the OCR engine and language data are downloaded from the jsDelivr CDN and cached; your image itself is never sent anywhere. In DevTools → Network you will see those engine and language files load, but no request carrying your image.

Can it read handwriting?↓

Only to a limited degree. The engine is trained on printed text, so neat block letters may be recognized, but cursive and messy handwriting usually produce poor results.

How can I improve OCR accuracy?↓

Use a sharp, high-resolution image with dark text on a light background, keep the text straight, crop away non-text areas, and select the correct language. For photos of paper, turn on Improve accuracy for photos.

Last updated: