Extract and Verify Book Titles from Bookshelf Photos using GPT-4o and Google Books

This n8n workflow automates the process of extracting book titles from an image of a bookshelf, verifying them against the Google Books API, and providing a structured response. It leverages the power of OpenAI's GPT-4o for image analysis and the Google Books API for data validation.

What it does

This workflow performs the following steps:

1. Receives an Image URL: It starts by listening for an incoming webhook request that contains the URL of an image (presumably a bookshelf photo).
2. Prepares Data for OpenAI: It takes the image URL and formats it into a prompt suitable for the OpenAI GPT-4o model, instructing it to identify and list all visible book titles.
3. Extracts Titles with GPT-4o: It sends the prepared prompt and image URL to the OpenAI API (using the gpt-4o model) to extract a list of potential book titles from the image.
4. Processes Extracted Titles: It then takes the raw output from GPT-4o and extracts the individual book titles, preparing them for verification.
5. Verifies Titles with Google Books: For each extracted title, it makes a request to the Google Books API to search for the book, attempting to validate its existence and retrieve additional information.
6. Responds to Webhook: Finally, it compiles the results (extracted titles and their verification status from Google Books) and responds to the initial webhook request with a structured JSON output.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• OpenAI API Key: An API key for OpenAI with access to the gpt-4o model.
• Google Books API Key: An API key for the Google Books API.
• Webhook Trigger: An external system or application capable of sending an HTTP POST request with an image URL to the n8n webhook.

Setup/Usage

1. Import the Workflow:
   • Copy the provided JSON code.
   • In your n8n instance, click "New" in the top left, then "Import from JSON".
   • Paste the JSON and click "Import".
2. Configure Credentials:
   • Locate the "OpenAI" node and configure your OpenAI API key credential.
   • Locate the "HTTP Request" node (for Google Books) and configure your Google Books API key. This might involve adding it as a query parameter or header depending on your Google Books API setup.
3. Activate the Webhook:
   • The "Webhook" node will display a URL once the workflow is saved and activated. This is the endpoint you will send your image URLs to.
4. Test the Workflow:
   • Send an HTTP POST request to the webhook URL. The request body should contain a JSON object with a key (e.g., imageUrl) holding the URL of a bookshelf photo.
   • Example Request Body:

     {
       "imageUrl": "https://example.com/path/to/your/bookshelf_photo.jpg"
     }
     

   • Observe the execution in n8n and the response received by your client.