n8n-nodes-disruptech-ocr

Extract text from images and PDFs — including scanned (image-based) PDFs — in your n8n workflows. This community node provides OCR (Optical Character Recognition) for images and PDFs using Tesseract.js, with automatic fallback for scanned documents where text cannot be selected or copied.

Built by Disrup-tech.com

Features

• OCR from Images — Extract text from PNG, JPG, TIFF, BMP, and other image formats using Tesseract.js
• Extract Text from PDFs — Pull text content from native text-based PDF documents
• Scanned PDF OCR Fallback — Automatically detect and OCR image-based PDFs (scanned documents) where normal text extraction returns nothing
• Multi-language Support — OCR supports 100+ languages via Tesseract language packs
• No External APIs — All processing happens locally, no data leaves your server

Installation

Via n8n Community Nodes (Recommended)

1. Open your n8n instance
2. Go to Settings → Community Nodes
3. Click Install a community node
4. Enter: n8n-nodes-disruptech-ocr
5. Click Install
6. Restart n8n when prompted

Via npm (Self-hosted)

cd ~/.n8n/nodes
npm install n8n-nodes-disruptech-ocr
# Restart n8n

Docker

Mount the node into your n8n container:

docker run -it --rm \
  --name n8n \
  -p 5678:5678 \
  -e N8N_CUSTOM_EXTENSIONS="/home/node/.n8n/custom/n8n-nodes-disruptech-ocr" \
  -v n8n_data:/home/node/.n8n \
  docker.n8n.io/n8nio/n8n

Usage

OCR from Image

Extract text from images using Tesseract OCR.

1. Add Baddak OCR node to your workflow
2. Set Operation to OCR from Image
3. Configure:
   • Input Binary Field: Name of the binary property containing the image (default: data)
   • Language: Tesseract language code (default: eng)

Example workflow:

[Read Binary File] → [Baddak OCR] → [Set Node]

Supported image formats: PNG, JPG/JPEG, TIFF, BMP, GIF, WebP

Language codes:

• eng - English
• ara - Arabic
• deu - German
• fra - French
• spa - Spanish
• chi_sim - Chinese (Simplified)
• jpn - Japanese
• Multiple languages: ara+eng, eng+deu+fra

Extract Text from PDF

Extract text content from PDF documents.

1. Add Baddak OCR node to your workflow
2. Set Operation to Extract Text from PDF
3. Configure:
   • Input Binary Field: Name of the binary property containing the PDF (default: data)
   • OCR Fallback for Scanned Pages: Enable this if the PDF is a scanned document (image-based) where text cannot be selected
   • OCR Language: Language for OCR fallback (default: eng)

Example workflow:

[HTTP Request (PDF URL)] → [Baddak OCR] → [Code Node]

Output:

{
  "text": "Extracted text content...",
  "pages": 5,
  "ocr": false
}

When OCR fallback is triggered on a scanned PDF:

{
  "text": "OCR'd text from scanned pages...",
  "pages": 3,
  "ocr": true
}

Examples

Basic Image OCR

1. Use Read Binary File to load an image
2. Connect to Baddak OCR with operation OCR from Image
3. Output contains text, confidence, and words count

Scanned PDF (Image-based PDF)

1. Use HTTP Request or Read Binary File to load the scanned PDF
2. Connect to Baddak OCR with operation Extract Text from PDF
3. Enable OCR Fallback for Scanned Pages
4. Set the correct OCR Language if not English
5. Output contains OCR'd text and ocr: true

Process PDF and Send via Email

1. HTTP Request — Download PDF from URL
2. Baddak OCR — Extract text (operation: Extract Text from PDF)
3. Send Email — Include extracted text in email body

Troubleshooting

Node not appearing after installation

• Restart your n8n instance
• Check the n8n logs for any errors

Low OCR accuracy

• Use higher resolution images (300 DPI recommended)
• Ensure good contrast between text and background
• Specify the correct language code
• Pre-process images to remove noise if needed

PDF extraction returns empty text

• The PDF may contain scanned images instead of selectable text
• Enable OCR Fallback for Scanned Pages in the node settings

License

MIT

Links

• Disrup-tech.com
• npm Package
• n8n Community Nodes