広告掲載
MCP.so
ログイン

MCP Image Recognition Server

@mario-andreschak

リポジトリ

基本情報

カテゴリ

メディアとデザイン

ライセンス

MIT

トランスポート

stdio

作者

mario-andreschak

MCP Image Recognition Server について

An MCP server that provides image recognition 👀 capabilities using Anthropic and OpenAI vision APIs

MCP Image Recognition Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "mcp-image-recognition": {
      "command": "python",
      "args": [
        "-m",
        "image_recognition_server.server"
      ]
    }
  }
}

MCP Image Recognition Server のツール

4

Base64-encoded image data and MIME type

Detailed description of the image

Path to an image file

Detailed description of the image

概要

What is MCP Image Recognition Server?

An MCP server that provides image recognition capabilities using Anthropic and OpenAI vision APIs. It supports multiple image formats and optional OCR text extraction via Tesseract.

How to use MCP Image Recognition Server?

Clone the repository, copy .env.example to .env, set API keys and preferences, then build and run via python -m image_recognition_server.server or run.bat server. Use tools describe_image (base64 input) and describe_image_from_file (file path). Environment variables control provider, model, OCR, and logging.

Key features of MCP Image Recognition Server

  • Image description using Claude Vision or GPT-4 Vision
  • Support for JPEG, PNG, GIF, WebP formats
  • Configurable primary and fallback provider (Anthropic/OpenAI)
  • Base64 and file-based image input
  • Optional Tesseract OCR text extraction
  • OpenRouter support for additional models

Use cases of MCP Image Recognition Server

  • Automatically describing images in chat or workflow applications
  • Extracting text from images using optional OCR
  • Integrating vision capabilities into MCP-powered tools
  • Reliable image analysis with fallback provider

よくある質問

1

What are the runtime requirements?

Python 3.8 or higher is required. Tesseract OCR is optional and only needed for text extraction.

2

How do I configure the vision provider?

Set `VISION_PROVIDER` to `anthropic` or `openai` in the `.env` file. Optionally set `FALLBACK_PROVIDER` for a secondary provider.

3

Can I use models other than the defaults?

Yes, via OpenRouter: set `OPENAI_BASE_URL` to `https://openrouter.ai/api/v1`, `OPENAI_MODEL` to the desired model (e.g., `anthropic/claude-3.5-sonnet:beta`), and `VISION_PROVIDER` to `openai`.

4

Does the server support Docker?

Yes. Build the image with `docker build -t mcp-image-recognition .` and run with `docker run -it --env-file .env mcp-image-recognition`.

5

What image formats are supported?

JPEG, PNG, GIF, and WebP are supported for both base64 and file-based input.

コメント

「メディアとデザイン」の他のコンテンツ

ProfessionalHeadshot.io

ProfessionalHeadshot.io

Official LinkedIn, Slack and GitHub profile photo specs with sources, headshot style guides by use and industry, and a free 0-100 headshot audit. No API key needed.

PixelDojo

PixelDojo

Generate images, video, audio and short films with 140+ AI models from any MCP client. PixelDojo picks the right model (Flux, Seedream, Seedance, Veo, Kling, WAN, Nano Banana, GPT Image), renders it and returns download

EveryGen AI MCP Server

EveryGen AI

EveryGen AI MCP Server lets AI assistants create videos, images, voiceovers, music and complete multi-shot films. Connect your EveryGen AI account, describe what you want, review the credit quote and approve the creation

Kinetune

Kinetune

Turn your songs into lyric videos (9:16, 16:9, 1:1) and Spotify Canvas loops. Find a song, browse Looks, get an exact credit quote, and render only after you say yes.

Designesy

LE-VAI

Design quality gate for AI-generated UI. Scores any live URL against a published contract of deterministic checks, with an A–F grade, a per-check breakdown and AI-slop pattern findings; validates DTCG token files and Lot

広告掲載