Skip to main content

SAAM - Document Analysis - Execute Batch OCR Process

ACCESS THE CUSTOM STEP FOLDER

SAAM - Document Analysis - Execute Batch OCR Process

Description​

This custom step is provided to enable point-and-click usage of the functionality available as part of the SAS Document Analysis offering from within the SAS Studio interface.

Use the SAAM - Document Analysis - Execute Batch OCR Process step in order to turn analysis documents of the following file types: PDF, PNG, JPG, JPEG, TIF, TIFF, BMP & ZIP. After the document analysis has finished you can run the companion step SAAM - Document Analysis - Produce Usage Report to generate a report about the performed analysis.

Features​

  • Multiple OCR engines supported (MS & AWS)
  • Parallelized batch-based end-to-end execution
  • Automated file conversion (between image and PDF formats)
  • Process tracking and entity mapping
  • Viya-ready outputs produced

User Interface​

  • SAAM - Document Analysis - Execute Batch OCR Process - Options Page​

SDA execute OCR batch process options page

Requirements​

  • SAS Viya 2025.10 or later
  • A license for SAS Document Analysis is required

Settings​

For more information about the different settings please refer to the SAS documentation linked below.

Documentation​

Change Log​

SAAM - Document Analysis - Execute Batch OCR Process​

  • Version 1.11 (13OCT2026)

    • Show the AI-generated content disclaimer at the top of the Options page, for every OCR engine

  • Version 1.10 (21AUG2026)

    • Add optional setting to assign a deterministic batch output folder name:
      • Generates a prefixed, timestamped name and passes it to the batch as BATCH_OUTPUT_FOLDER_NAME
      • Reports the resulting batch output directory in the results
      • Retains _sda_batch_output_folder_name and _sda_batch_output_directory for downstream steps

  • Version 1.9 (28MAY2026)

    • Renamed step from "OCR - Document Analysis - Execute Batch OCR Process" to "SAAM - Document Analysis - Execute Batch OCR Process" to conform to the SAAM naming standard

  • Version 1.8 (22APR2026)

    • Add the following options and associated defaults:
      • Retain line breaks (True)
      • Generate images from PDFs (False)
      • Output mapping file format (XSLX)

  • Version 1.7 (22APR2026)

    • Remove "bypass SSL" option. Default to False

  • Version 1.6 (16APR2026)

    • Communicate exit codes for failed SDA batch processes

  • Version 1.5 (19FEB2026)

    • Use absolute path for python command

  • Version 1.4 (04NOV2025)

    • Add PaddleOCR to OCR engine options
    • Add MS OCR model name - available when "Microsoft" is selected as the OCR engine
    • Add Custom Document Intelligence model name - available when "Custom" is selected as the MS OCR model
    • Remove optional setting to maintain directory structure

  • Version 1.3 (25FEB2025)

    • Use launcher context name instead ID

  • Version 1.2 (21NOV2024)

    • Added option to enable syncronous processing
    • Added option to suppress SSL check

  • Version 1.1 (09OCT2024)

    • Implemented feedback

  • Version 1.0 (07OCT2024)

    • Initial version