Skip to content

Read Words Activity

Overview

The Read Words activity extracts text content from a specified image using Optical Character Recognition (OCR). It transfers the recognized text, line by line, into a designated variable in JSON format. Optionally, you can define a specific region within the image for text extraction.

Parameter Table

Field Description Requirement
File Name The full path and name of the image file (e.g., .png, .jpg) to be processed. Required
X-coordinate The X-coordinate (horizontal position) of the top-left corner of the region to read within the image. Must be used in conjunction with Y-coordinate, Width, and Height. Optional
Y-coordinate The Y-coordinate (vertical position) of the top-left corner of the region to read within the image. Must be used in conjunction with X-coordinate, Width, and Height. Optional
Width The width of the region to read within the image. Must be used in conjunction with X-coordinate, Y-coordinate, and Height. Value must be an integer. Optional
Height The height of the region to read within the image. Must be used in conjunction with X-coordinate, Y-coordinate, and Width. Value must be an integer. Optional
Language The language library to use for Optical Character Recognition (OCR). Supported options include TUR, ENG, RUS, OCR-B and NLD. Required
Result Text Name The name of the variable where the OCR result, in JSON format, will be stored. Required

Action Types & Examples

OCR Result JSON

  • Format: string (JSON)
  • Example Result: {"result":[{"lineNumber":1, "words":[{"text":"Hello","x":10,"y":10,"width":200,"height":20}, {"text":"World","x":220,"y":10,"width":205,"height":20}]}]}

OCR Result JSON with Confidence Score

  • Format: string (JSON)
  • Example Result: {"result":[{"words":[{"confidence":81,"x":2,"width":53,"y":4,"text":"Success","height":17}, {"confidence":78,"x":59,"width":97,"y":4,"text":"Ratio","height":17}],"lineNumber":1}]}

Implementation Examples

Field Setup

  • File Name: C:\Robusta\document.png
  • Language: ENG
  • Result Text Name: ocrOutput

Execution Parameters

  • File Name: C:\Robusta\invoice.png
  • X-coordinate: 50
  • Y-coordinate: 100
  • Width: 300
  • Height: 50
  • Language: TUR
  • Result Text Name: invoiceHeader

Technical Notes

When defining a specific region for text extraction, all four coordinate parameters (X-coordinate, Y-coordinate, Width, Height) must be provided. If any of these are left blank, all four must be blank, in which case the entire image will be processed. The Width and Height values must be integers. The OCR result, including text and optional confidence scores, is transferred to the specified variable in JSON format.