Omikuji Tools Image & Video

Image to Text (OCR)

Turn text in images, screenshots, photos and PDFs into editable text. Japanese (horizontal and vertical) and English. Your images never leave this device.

Features

  • Extract text from images, screenshots, photos and PDFs – Japanese (horizontal and vertical) and English
  • Paste with Ctrl+V, capture the screen, use your phone camera, or batch several images and every PDF page
  • Read only a selected area; small images are upscaled, converted to grayscale and dark backgrounds inverted automatically
  • Cleans up Japanese spacing and joins broken lines; uncertain words are highlighted for review

Use cases

Copy text from screenshots

Grab text from images and app screens that cannot be selected.

Paper documents

Photograph a document and turn it into searchable text.

Vertical Japanese

Read books and notices set in vertical writing.

Scanned PDFs

Convert pages you cannot select into text.

Privacy

Images and PDFs are processed only inside your browser and are never uploaded. On first use the OCR engine (about 4 MB) and language data (2–3 MB per language) are downloaded from jsDelivr (cdn.jsdelivr.net); your images are not sent. Language data is stored in your browser for next time.

FAQ

Are my images uploaded?

No. Text recognition (tesseract.js) runs entirely in your browser. The only network access is downloading the engine and language data from jsDelivr the first time.

Can it read vertical Japanese?

Yes. Set the direction to Vertical. For complex multi-column layouts, select one column at a time for better accuracy.

How can I improve accuracy?

Straight, large and sharp text works best. Select only the area you need, choose the right language, and use grayscale for photos. Handwriting is not recognised well.

Can it read PDFs?

Yes. Scanned PDFs whose text cannot be selected are rendered page by page and read (up to the first 30 pages).

Related tools

SVG Converter Image Merge Batch Watermark Image & Print Size Calculator LocalCanvas PhotoBatch Image Converter WebP Converter EXIF Cleaner Local Mask Image to PDF PDF to Image

Add images

Drop images or PDFs here, or click to choose

Paste a screenshot with Ctrl+V (⌘V). PNG, JPEG, WebP, GIF, BMP or PDF; several at once is fine

Settings

Language
Text direction

How to use

  1. Drop images or PDFs, or paste a screenshot with Ctrl+V (on a phone, tap Take a photo).
  2. To read only part of an image, turn on area selection and adjust the frame.
  3. Choose the language and direction, then press Extract text. The first run downloads 2–3 MB of language data.
  4. Check the highlighted words, fix them in the text box, then copy or save.

Privacy

Images and PDFs are processed only inside your browser and are never uploaded. On first use the OCR engine (about 4 MB) and language data (2–3 MB per language) are downloaded from jsDelivr (cdn.jsdelivr.net); your images are not sent. Language data is stored in your browser for next time.