This project provides a high-performance OCR API service using FastAPI and DdddOcr, targeting developers needing to integrate image text recognition, sliding captcha matching, and object detection into their applications. It offers a simple, Docker-friendly deployment for quick integration.

How It Works

The API leverages the DdddOcr library to perform OCR tasks. It exposes three main endpoints: /ocr for general text recognition, /slide_match for comparing images in sliding captcha puzzles, and /detection for object detection within images. Requests can be made using either image files or Base64 encoded strings, with options to control output verbosity and image processing.

Quick Start & Requirements

• Install/Run:
  • Clone the repository: git clone https://github.com/your-repo/ddddocr-api.git
  • Navigate to the directory: cd ddddocr-api
  • Build and run with Docker:

    docker build -t ddddocr-api .
    docker run -d -p 8000:8000 --name ddddocr-api-container ddddocr-api
    

  • Alternatively, run directly: python app/main.py or uvicorn app.main:app --reload
• Prerequisites: Docker (20.10+), Docker Compose (1.29+). Recommended OS: Linux (Ubuntu 20.04 LTS+).
• Verification: Access http://localhost:8000/docs for Swagger UI.
• Links: GitHub Repository (Note: The provided README uses a placeholder URL for cloning).

Highlighted Details

• Supports OCR, sliding captcha matching, and object detection.
• Accepts image input via file upload or Base64 encoding.
• Provides options for probability output and PNG fixing in OCR.
• Includes example API calls for Python, Node.js, C#, PHP, and Go.

Maintenance & Community

• The project is maintained by sml2h3.
• No specific community links (Discord/Slack) or roadmap are mentioned in the README.

Licensing & Compatibility

• License: MIT License.
• Compatibility: The MIT license is permissive and generally compatible with commercial and closed-source applications.

Limitations & Caveats

• The README uses a placeholder URL for cloning the repository.
• Production deployment recommendations include HTTPS and authentication, which are not configured by default.