HanvonScanner OCR MCP

English | Chinese

What is the HanvonScanner OCR MCP Service

HanvonScanner is a general document recognition and image processing cloud service launched by Beijing Zhongke Yueshen Technology. It is fully compatible with MCP protocol clients, providing out-of-the-box document processing capabilities for AI applications, agents, and automation systems.

• OCR Intelligent Recognition Service

  Provides professional-grade text recognition, layout analysis, and structured information extraction. It supports the automatic extraction of key information from various documents such as ID cards, bank cards, invoices, and contracts, while completely restoring the original document layout. Its capabilities cover special elements like Chinese and English handwriting, complex formulas, seals, and charts, and it supports over 52 printed languages. This significantly enhances the data processing efficiency of large models in scenarios like document understanding, content generation, and intelligent Q&A.

• AI Image Processing Service

  Integrates advanced computer vision technology to provide core capabilities such as intelligent document detection, AI enhancement, high-definition processing, screen moire removal, handwriting removal, watermark removal, and line art extraction. It can automatically correct and optimize blurred or skewed documents taken by mobile phones, converting them into high-definition scanned copies with a single click. It also supports the precise removal of interfering information like handwritten annotations and watermarks, purifying line art and document content to meet various document digitization and restoration needs.

Get API_KEY

Go to the HanvonScanner Open Platform, log in, and register to obtain an API_KEY. The MCP service is currently supported for integration into agents and workflows.

NPX Installation

{
  "mcpServers": {
    "hanvonsc-mcp": {
      "command": "npx",
      "args": ["-y", "@hanvonsc/mcp-server"],
      "env": {
        "API_KEY": "<API_KEY>"
      },
      "timeout": 60000
    }
  }
}

HanvonScanner OCR MCP Features

• doc-ocr: General Text Recognition

  • Performs text recognition on images/PDFs, supporting Chinese and English handwriting/printed text, formulas, etc.;
  • Input:
    • filePath (string, choose one between this and base64): File path of the image or PDF document
    • base64 (string, choose one between this and filePath): Base64 encoding of the image or PDF document
  • Output: Structured JSON string after recognition
  • Supported formats:
    • png, jpg, jpeg, pdf

• doc-layout-ocr: General Document Layout Analysis

  • Performs layout analysis on images/PDFs, supporting the restoration of tables, formulas, images, seals, headers/footers, and column information; supports handwriting/printed text; supports languages such as Chinese, English, Japanese, and Korean;
  • Input:
    • filePath (string, choose one between this and base64): File path of the image or PDF document
    • base64 (string, choose one between this and filePath): Base64 encoding of the image or PDF document
  • Output: Formatted data
  • Supported formats:
    • png, jpg, jpeg, pdf

• file-to-word: Convert and Merge PDF Documents and Images into Word Documents

  • Converts and merges PDF documents and images into a Word document
  • Input:
    • filePaths (string[], required): File paths, passed as an array of file paths
  • Output:
    • uri: File download address
  • Supported formats:
    • pdf, png, jpg, jpeg

• file-to-excel: Convert and Merge PDF Documents and Images into Excel Documents

  • Converts and merges PDF documents and images into an Excel document
  • Input:
    • filePaths (string[], required): Paths to PDF documents and images, passed as an array of file paths
  • Output:
    • uri: File download address
  • Supported formats:
    • pdf, png, jpg, jpeg

• image-process: Image Processing

  • Provides: Document detection, image orientation correction based on text direction, AI enhancement, handwriting removal, watermark removal, AI high-definition, screen moire removal, and line art extraction;
  • Input:
    • base64img: Choose one between this and filePath, base64 encoding of the image
    • filePath: Choose one between this and base64img, file path of the image
    • filter: Filter type 0: No filter 1: AI enhancement 2: AI high-definition 3: Screen moire removal 4: Ink saving 5: Black text on white background 6: Line art extraction, default: 0
    • deHw: Whether to remove handwriting 1: Yes 0: No, default: 0
    • deWm: Whether to remove watermark 1: Yes 0: No, default: 0
    • outputType: Output type 0: jpg 1: pdf, default: 0
  • Output:
    • Image or PDF
  • Supported formats:
    • pdf, png, jpg, jpeg

License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, provided you comply with the terms and conditions of the MIT License.