Brazilian CNH (Driver's License) OCR API

The Complexity of Brazilian Identity Documents

Processing the Brazilian Carteira Nacional de Habilitação (CNH) is notoriously difficult for standard text-recognition tools. Unlike standardized North American or European IDs, the CNH contains highly localized data structures. Critical identifiers like the CPF (Tax ID) and RG (Identity number) are grouped alongside expansive fields for 'Filiação' (Mother and Father's names) and the 'Espelho' (card security number). Generic models frequently mislabel these fields or fail to parse the DD/MM/YYYY date formats accurately. To see how our dedicated identity models overcome these regional challenges, explore our comprehensive driver license OCR capabilities.

Edge-Routed Precision & Seamless Integration

Our CNH OCR engine is engineered to understand the exact topography of Brazilian licenses. We provide this as a pure REST API—meaning you don't need to bloat your mobile app or frontend with heavy SDKs, though lightweight wrappers are available for multiple languages. Because our infrastructure is deployed on Cloudflare edge networks, authentication and routing happen instantly, guaranteeing ultra-low latency for user requests originating in South America. The API natively accepts direct file uploads, Base64-encoded strings, or image URLs, providing absolute flexibility for your engineering team.

High-Growth Use Cases in LatAm

Integration & AI Prompts

Ready to build? You can pass the document image directly via a Base64 string or file upload. For deep technical integration patterns, check our developer documentation. Try the prompt below in your AI coding assistant.

import requests
import base64

# Prepare Base64 Image of the Brazilian CNH
with open("cnh_sample.jpg", "rb") as image_file:
    base64_image = base64.b64encode(image_file.read()).decode('utf-8')

url = "https://api.structocr.com/v1/driver-license"
headers = {
    "x-api-key": "YOUR_API_KEY",
    "Content-Type": "application/json"
}

# Our API handles pure Base64 strings natively
payload = {
    "img": base64_image 
}

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

if result.get('success'):
    data = result['data']
    print(f"CPF Extracted: {data['personal_number']}")
    print(f"License Number: {data['document_number']}")

{
  "success": true,
  "data": {
    "type": "drivers_license",
    "country_code": "BRA",
    "region": "SP",
    "document_number": "12345678900",
    "personal_number": "111.222.333-44",
    "surname": "SILVA SANTOS",
    "given_names": "MARIA LUIZA",
    "sex": null,
    "date_of_birth": "1985-10-15",
    "place_of_birth": "SAO PAULO/SP",
    "date_of_issue": "2023-01-20",
    "date_of_expiry": "2028-01-20",
    "date_of_first_issue": "2005-06-10",
    "issuing_authority": "DETRAN SP",
    "address": null,
    "vehicle_class": "B",
    "extra_details": {
      "fathers_name": "JOAO SILVA",
      "mothers_name": "ANA SANTOS",
      "rg_number": "1234567 SSP SP",
      "card_security_number": "0000000000"
    }
  }
}