MCP Image Recognition

🚀 MCP Image Recognition Server

An MCP server that uses Anthropic and OpenAI Vision APIs to provide image recognition capabilities. Version 0.1.2.

🚀 Quick Start

The MCP Image Recognition Server leverages Anthropic and OpenAI Vision APIs to offer robust image recognition features. It supports multiple image formats and provides flexible configuration options.

✨ Features

• Use Anthropic Claude Vision or OpenAI GPT - 4 Vision for image description.
• Support multiple image formats (JPEG, PNG, GIF, WebP).
• Configurable primary and backup providers.
• Support image input in Base64 and file formats.
• Optional Tesseract OCR text extraction feature.

📦 Installation

Dependencies

• Python 3.8 or higher.
• Tesseract OCR (optional) for text extraction:
  • Windows: Download and install from [UB - Mannheim/tesseract](https://github.com/UB - Mannheim/tesseract/wiki).
  • Linux: sudo apt - get install tesseract - ocr.
  • macOS: brew install tesseract.

Steps

1. Clone the repository:

git clone https://github.com/mario - andreschak/mcp - image - recognition.git
cd mcp - image - recognition

2. Create and configure the environment file:

cp.env.example.env
# Edit the.env file to set API keys and preferences.

3. Build the project:

build.bat

💻 Usage Examples

Running the Server

Start the server using Python:

python -m image_recognition_server.server

Start the server via the batch script:

run.bat server

Run the server in development mode (with MCP inspector):

run.bat debug

Available Tools

1. describe_image

   • Input: Base64 - encoded image data and MIME type.
   • Output: Detailed description of the image.

2. describe_image_from_file

   • Input: Image file path.
   • Output: Detailed description of the image.

📚 Documentation

Environment Configuration

• ANTHROPIC_API_KEY: Anthropic API key.
• OPENAI_API_KEY: OpenAI API key.
• IMAGE_SIZE: Size of the processed image (default is "256x256").
• MAX_ITERATIONS: Maximum number of iterations (default is 100).

OpenRouter Instructions

• If using OpenRouter, refer to its documentation for detailed configuration information.
• Ensure that the correct API key and model endpoint are set.

Default Models

• Anthropic's Claude Vision model.
• OpenAI's GPT - 4 Vision model.

🔧 Technical Details

Testing

Run all tests:

run.bat test

Debugging

Run the server in debug mode:

run.bat debug

📄 Release History

• Version 0.1.2:

  • Fixed known compatibility issues.
  • Improved OCR error handling and added comprehensive test coverage for the OCR feature.

• Version 0.1.1:

  • Initial release, including basic features and documentation support.