Platform

Platform Side Services

Selected supporting services

Use one focused part of the document platform when the complete extraction workflow is more than the job requires.

How the side services prepare and read documents
  1. Prepare

    Convert, render, assemble, merge, or split

  2. Map

    Locate and classify regions on the page

  3. Read

    Recognize text and page observations

Conversion & PDF Handling

Current service

Prepare the source before extraction begins.

Move common office, markup, image, and PDF sources into consistent pages or documents for the next step.

  • Convert supported office, markup, and image sources to PDF
  • Render PDFs and mixed files into page images
  • Assemble images into PDFs
  • Merge and split PDF documents

Layout Classification

Current service

Know what is on the page and where it sits.

Locate page regions, assign structural labels, and classify pictures before deeper extraction or review.

  • Bounding boxes, labels, and confidence
  • Titles, text, tables, forms, checkboxes, and figures
  • Region hierarchy and document-level cleanup
  • Detailed visual classification for picture regions

Optical Text Recognition

Current service

Read the page without running the complete document model.

Recognize page text and return words, lines, geometry, language, confidence, and richer native observations when requested.

  • Recognized text, words, and lines
  • Coordinates, language, and confidence
  • Text-first and richer layout reading paths
  • Focused output for search, review, or downstream processing