Skip to main content

​
Overview

Extract Images is a feature that extracts all images from unstructured data sources, including graphics, plots, tables, charts, diagrams, and other visual content. This feature processes Corvic Tables produced by the Document Digitization feature to extract and organize images with their metadata for further processing and analysis.

​
Category

Corvic Tables - This feature is designed to work with Corvic Tables, specifically those produced by Document Digitization, enabling you to extract and organize visual content from digitized documents.

​
Input

Corvic Table - The Extract Images feature accepts a Corvic Table produced by the Document Digitization feature as input. The input Corvic Table should contain digitized markdown content with embedded image references.
The input must be a Corvic Table created using the Document Digitization feature. This ensures the table contains the necessary structure and metadata for image extraction.
Learn more: Document Digitization - Convert unstructured data to Markdown format

​
Output

Corvic Table - The Extract Images feature produces a new Corvic Table containing all extracted images and their metadata. The output includes:
  • Extracted Images: All images found in the input data including graphics, plots, tables, charts, diagrams, and other visual content
  • Image Metadata: Information about each image including dimensions, format, position, and context
  • Structured Organization: Images organized in a format ready for further processing, analysis, or embedding
The output Corvic Table provides a structured representation of all visual content extracted from your digitized documents, making it easy to process images separately, generate image embeddings, or use them in downstream workflows.

​
Parameters

​
Usage Example

To use Extract Images in a Workflow:
  1. Add your Corvic Table (from Document Digitization) to the Workflow canvas
  2. Click the ”+” button next to the Corvic Table
  3. Select “Extract Images” from the actions menu
  4. Select the input Corvic Table (if not already selected)
  5. Optionally provide a name for the output Corvic Table
  6. Run the Workflow to execute the image extraction
  7. Review the generated Corvic Table containing all extracted images and metadata

Document Digitization

Learn how to digitize documents to create Corvic Tables with image references.

Workflows

Learn how to build workflows using Workflows.

Corvic Tables

Understand how Corvic Tables work and how to create them.

Embed

Generate embeddings from extracted images using image semantic embeddings.