Germany Personalausweis OCR Python SDK

Parsing Personalausweis Challenges

The German Personalausweis presents unique challenges for OCR. Firstly, the presence of specific German characters (Umlauts: ä, ö, ü and Eszett: ß) requires highly accurate native script recognition. Secondly, ensuring the authenticity of the document requires reliably locating and strictly validating the Machine-Readable Zone (MRZ) on the back of the card, while correctly handling ICAO transliterations between the visual text and the MRZ.

Why StructOCR for Germany

Our model is specifically trained on a large dataset of German Personalausweis documents, ensuring high accuracy for id parsing api applications. The Python SDK simplifies integration with a clean and intuitive API, allowing you to process documents with just a few lines of code. It accurately captures complex German typography and actively parses the embedded MRZ to automate KYC verification workflows.

Common Use Cases in Germany

• Digital Onboarding: Verify users for Fintech apps in Germany, automatically matching VIZ data against the MRZ for enhanced security.
• Telecom Registration: Automate SIM card registration with native support for German addresses and characters.
• Hotel Check-in: Speed up guest registration workflows with automated MRZ scanning.

Python SDK Integration

Install the SDK via pip: `pip install structocr`. The SDK automatically handles image preprocessing, maps visual fields, and extracts the raw TD1 MRZ lines.

Prerequisite: Python 3.6+ and `structocr` library installed.

from structocr import StructOCR

# Get 200 free credits:
# 👉 https://structocr.com/register
# Initialize with your API Key
client = StructOCR("YOUR_API_KEY_HERE")

def scan_germany_id():
    # Note: Supports JPG, PNG, WebP (Max 4.5MB)
    # Target: Personalausweis
    image_path = "germany_national_id.jpg"

    try:
        print(f"Scanning {image_path}...")
        
        # The SDK handles file upload and API communication
        result = client.scan_national_id(image_path)

        if result.get('success'):
            data = result['data']
            print("✅ Germany Extraction Successful!")
            
            # Basic Identity
            print(f"Region:      {data.get('country_code')}")
            print(f"Name:        {data.get('given_names')} {data.get('surname')}")
            print(f"ID Number:   {data.get('document_number')}")
            
            # Demographics
            print(f"DOB:         {data.get('date_of_birth')} ({data.get('sex')})")
            print(f"Address:     {data.get('address')}")
            
            # Extract TD1 MRZ Data
            additional = data.get('additional_fields', {})
            if additional.get('mrz_line_1'):
                print("\n🔍 MRZ Data Extracted & Validated:")
                print(f"Line 1: {additional.get('mrz_line_1')}")
                print(f"Line 2: {additional.get('mrz_line_2')}")
                print(f"Line 3: {additional.get('mrz_line_3')}")
        else:
            print(f"❌ Extraction Failed: {result.get('error')}")

    except Exception as e:
        print(f"An error occurred: {e}")

if __name__ == "__main__":
    scan_germany_id()

JSON Response Example

The SDK returns a Python dictionary matching this JSON structure, including the parsed visual data with native German characters and the raw Machine-Readable Zone (MRZ).

{
  "success": true,
  "data": {
    "type": "national_id",
    "country_code": "DEU",
    "nationality": "DEUTSCH",
    "document_number": "L01X00T2Z",
    "card_series": "",
    "personal_number": "",
    "surname": "MÜLLER",
    "given_names": "MICHAEL",
    "sex": "M",
    "date_of_birth": "1990-05-15",
    "place_of_birth": "BERLIN",
    "address": "Unter den Linden 5, 10117 Berlin",
    "date_of_issue": "2020-01-01",
    "date_of_expiry": "2030-01-01",
    "issuing_authority": "Bundesdruckerei",
    "additional_fields": {
      "tramite_number": null,
      "mrz_line_1": "IDDEUL01X00T2Z5<<<<<<<<<<<<<<<",
      "mrz_line_2": "9005156M3001018D<<<<<<<<<<<<<4",
      "mrz_line_3": "MUELLER<<MICHAEL<<<<<<<<<<<<<<"
    }
  }
}