0-CodeKit

A toolbox of no-code utilities

Actions40

Overview

The AI - Picture Text Recognition operation in this custom n8n node allows you to extract text from images by providing the URL of the image. This is commonly known as Optical Character Recognition (OCR). It is useful for automating workflows that require reading and processing text from scanned documents, receipts, screenshots, or any other images containing textual information.

Practical examples:

Automatically extracting invoice numbers from uploaded receipt images.
Reading addresses from scanned postal mail.
Digitizing printed forms for further automated processing.

Properties

Name	Meaning
URL	URL of the image you want to analyse

Output

The output will be a JSON object containing the results of the text recognition process. While the exact structure depends on the API response, you can expect fields such as:

The recognized text extracted from the image.
Additional metadata about the recognition process (e.g., confidence scores, bounding boxes).

Example output:

{
  "text": "Recognized text from the image",
  "confidence": 0.98,
  "language": "en"
}

Note: Actual fields may vary depending on the underlying service.

Dependencies

External Service: Requires access to the CodeKit API for AI operations.
Credentials: You must configure the codeKitApi credential in n8n for authentication.

Troubleshooting

Common Issues:
- Invalid Image URL: If the provided URL is incorrect or inaccessible, the node may return an error or empty result.
- Unsupported Image Format: Only standard image formats (JPEG, PNG, etc.) are supported.
- API Key Missing/Invalid: If the codeKitApi credential is not set up correctly, authentication errors will occur.
Error Messages & Resolutions:
- "error": "Request failed with status code 401"
  Resolution: Check your API credentials in n8n.
- "error": "Image not found"
  Resolution: Verify the image URL is correct and publicly accessible.
- "error": "Unsupported file type"
  Resolution: Use a supported image format.