OCR

Overview

The OCR (Optical Character Recognition) integration can be used to convert documents and images (such as invoices, reports, and forms) into machine-readable text. This extracted text can then be used to create a document processing system that streamlines operations.

Use cases

Extract text from an image

Use an Input component coupled with an Image component to visualize an image in your application. Connect a Button component to the OCR integration and quickly extract text from the provided URL to pass through to subsequent APIs or Workflows.

Extract text from an image of a check and use Python to structure it in json format

Creating OCR steps

Actions

From File

Given an image or PDF file, convert to machine-readable text

Field	Description
File	A file to be converted to text. Supported file types: `['jpg', 'jpeg', 'png', 'bmp', 'pdf', 'tiff']`

info

If using a FilePicker component to submit files, the syntax will be {{FilePicker1.files[0]}}.

From URL

Given a URL to an image or PDF file, convert to machine-readable text

Field	Description
URL	A URL to a file to be converted to text. Supported file types: `['jpg', 'jpeg', 'png', 'bmp', 'pdf', 'tiff']`

Overview​

Use cases​

Extract text from an image​

Creating OCR steps​

Actions​

From File​

From URL​