Blab Document OCR icon

Blab Document OCR

Extract text from document images using Upstage Document OCR. Supports JPEG, PNG, BMP, PDF, TIFF, HEIC, DOCX, PPTX, XLSX, HWP, HWPX formats.
Join our community

Overview

This node performs Optical Character Recognition (OCR) on document images using the Upstage Document OCR API. It supports various file formats including JPEG, PNG, BMP, PDF, TIFF, HEIC, DOCX, PPTX, XLSX, HWP, and HWPX. The node is useful for extracting text and structured data from scanned documents, images, and other file types for automation, data entry, or content analysis workflows.

Use Case Examples

  1. Extract text from scanned invoices or receipts to automate data entry.
  2. Convert PDF documents into searchable text for archiving and retrieval.
  3. Analyze images containing text to extract words, pages, or confidence scores for quality control.

Properties

Name Meaning
Binary Property Name of the input item binary property that contains the file to be processed by OCR.
Model The OCR model to use for text extraction. The alias 'ocr' points to the latest stable model.
Schema Optional parameter specifying the response format to convert output to a corresponding OCR API format.
Return Determines what data to return from the OCR response.

Output

JSON

  • text - Extracted text from the document (when 'Text Only' return mode is selected).
  • pages - Array of pages with OCR data (when 'Pages Array' return mode is selected).
  • words - Array of words extracted from the document (when 'Words Array' return mode is selected).
  • confidence - Confidence score of the OCR along with model version and number of billed pages (when 'Confidence Score' return mode is selected).
    - Complete OCR response object from the Upstage API (when 'Full Response' return mode is selected).
  • error - Error message if OCR processing fails and continueOnFail is enabled.
  • statusCode - HTTP status code of the error response.
  • error_code - Error code returned by the OCR API or 'unknown_error' if not specified.
  • timestamp - Timestamp of the error occurrence.

Dependencies

  • Upstage Document OCR API
  • An API key credential for authentication

Troubleshooting

  • Common issues include missing or incorrectly named binary property containing the file, which results in an error 'No binary data found in property'.
  • File size exceeding 50MB limit triggers an error 'File size exceeds 50MB limit'.
  • Invalid or unexpected response format from the OCR API causes an error 'Invalid response format from Upstage OCR API'.
  • Network or authentication errors may occur if the API key credential is missing or invalid.
  • To resolve errors, ensure the binary property name matches the input data, file size is within limits, and valid API credentials are configured.

Links

Discussion