Full-Field USA DMV Vehicle Registration & Title OCR

The Challenge: Extreme Format Fragmentation

The United States presents the ultimate OCR challenge: fragmentation. There is no single national format. Instead, all 50 state DMVs issue their own distinct vehicle registration and Title documents, each with wildly different layouts, fonts, background colors, and terminology. Traditional OCR systems that rely on fixed bounding-box templates completely fail in the US market because it is impossible to maintain templates for every state variation and annual redesign. Learn more about our template-free capabilities on our Vehicle Registration OCR product page.

Template-Free Semantic Key-Value Mapping

StructOCR abandons outdated template matching. Our engine uses advanced Large Language Model (LLM) backed semantic Key-Value mapping to intelligently 'read' the document like a human does. Whether a state calls it a 'VIN', 'Vehicle ID', or 'Chassis Num', our system correctly maps it to the standard `vin` field. Universal specs are normalized perfectly, while unique state-specific data points are dynamically captured into our `localized.generic.unmapped_fields` object. Discover how this fits into your nationwide automotive workflows in our Vehicle Registration Solutions guide.

Core Industry Applications

Integration & API Documentation

Easily integrate our full-field OCR API. For full technical details, visit our Developer API Documentation.

import requests
import base64

# 📸 Test for free with your own image upload: https://structocr.com/developers/vehicle-registration-ocr

with open("california_dmv_registration.jpg", "rb") as image_file:
    base64_image = base64.b64encode(image_file.read()).decode('utf-8')

url = "https://api.structocr.com/v1/vehicle-registration"
headers = {
    "x-api-key": "YOUR_API_KEY",
    "Content-Type": "application/json"
}
payload = {
    "img": base64_image
}

response = requests.post(url, headers=headers, json=payload)
result = response.json()

if result.get('success'):
    data = result.get('data', {})
    std = data.get('standardized', {})
    print("✅ Standardized VIN:", std.get('vin'))
    print("✅ Standardized Make:", std.get('make_normalized'))
    print("✅ Standardized Year:", std.get('year'))
    
    # State specific details are captured in the generic fallback
    loc = data.get('localized', {})
    if 'generic' in loc and loc['generic'].get('unmapped_fields'):
        unmapped = loc['generic']['unmapped_fields']
        print("✅ State/Jurisdiction:", unmapped.get('state'))
        print("✅ Title Number:", unmapped.get('title_number'))

{
  "success": true,
  "data": {
    "document": {
      "country_code": "US",
      "document_type": "VEHICLE_REGISTRATION",
      "confidence_score": 0.99
    },
    "standardized": {
      "vin": "1G1RC6E45BU123456",
      "plate": {
        "number": "ABC1234",
        "formatted": "ABC 1234"
      },
      "make_normalized": "Chevrolet",
      "year": 2011
    },
    "localized": {
      "generic": {
        "unmapped_fields": {
          "state": "California",
          "title_number": "12345678",
          "expiration_date": "2024-12-31"
        }
      }
    }
  }
}