---
title: "OCR Data Extractor"
section: "Agent store — Documents — Data Extraction"
canonical_url: "https://leverge.ai/agents/documents/data-extraction/ocr-extractor"
category: "Documents"
process: "Data Extraction"
trigger: "Run on demand"
publisher: "Ailoitte Technologies Private Limited"
---

# OCR Data Extractor

Read a scanned invoice or receipt and pull out the header fields and line items as structured data.

## At a glance

- Business function: Documents
- Process: Data Extraction
- Runs: run on demand
- Run cost: 5 credits
- Uses a knowledge base: no

## What it needs

- **Scanned document** (file upload) — A scanned or photographed PDF. Pages are rasterized and read with OCR.
- **Extraction settings** (form)

## How it works

1. Rasterizing pages
2. Reading the scan
3. Checking scan quality
4. Checking the numbers add up

## What it returns

- **Checks** (validation result)
- **Header fields** (metadata grid)
- **Line items** (table of results)
- **Scan quality** (metadata grid)

---

Source: https://leverge.ai/agents/documents/data-extraction/ocr-extractor — Leverge