OCR Bro icon

OCR Bro

Extract text from images (OCR) or PDFs

Actions2

OCR from Image

AI-generated

Summary

Extract text from an image file provided as binary data using Tesseract OCR with optional multi-language support.

Inputs

  • Input Binary Field (required) — The name of the binary property containing the image file to be processed.
  • Language — Tesseract language code(s) for OCR, e.g., 'eng' for English. Multiple languages can be specified by separating codes with '+', such as 'eng+spa'.

Output shape

a list of items each containing JSON with the extracted text ('text'), confidence score ('confidence'), and number of words recognized ('words'), alongside the original binary data of the image

Processes each input item individually; throws an error if the specified binary property is missing or the file is not an image, unless 'Continue On Fail' is enabled, in which case errors appear in the output JSON. Uses Tesseract.js worker initialized with the specified language(s). Output includes recognition confidence and word count.

Links

Discussion