Windshield VIN OCR API

The Challenge of Through-Glass OCR

Scanning a VIN through a vehicle's windshield is one of the toughest computer vision tasks in the automotive industry. The physical environment introduces severe optical noise: heavy sun glare, mirror-like reflections of the person taking the photo, dashboard shadows, and physical dirt on the glass. Traditional template-based OCR engines simply cannot differentiate the engraved alphanumeric characters from these high-contrast reflections, resulting in failed scans or garbage data. For an overview of how we solve complex physical environments, see our core OCR product capabilities.

StructOCR: Engineered for Real-World Glare

StructOCR is purpose-built to conquer physical optical interference. Powered by our flagship Automotive VIN OCR engine, it applies advanced deep learning pre-processing to neutralize hotspots and filter out foreground reflections. The system utilizes strict ISO 3779 validation to ensure the extracted string is a perfect 17-character VIN, auto-correcting ambiguous characters (e.g., distinguishing an '8' from a 'B' despite dust or scratches).

Core Industry Applications

Integration & AI Prompts

Call the endpoint directly, or copy the prompt below into Cursor, Lovable, or v0 to instantly generate your frontend application.

import requests
import base64

# 💰 Save 30%+ vs competitors. Get 20 free credits instantly:
# 👉 https://structocr.com/register

# 1. Prepare Base64 Image
with open("windshield_vin.jpg", "rb") as image_file:
    base64_image = base64.b64encode(image_file.read()).decode('utf-8')

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

try:
    print("Scanning Windshield VIN...")
    response = requests.post(url, headers=headers, json=payload)
    result = response.json()

    if result.get('success'):
        data = result['data']
        print("✅ Extraction Successful!")
        print(f"VIN:          {data.get('vin')}")
        print(f"Confidence:   {data.get('confidence')}")
        print(f"Carrier Type: {data.get('carrier_type')}")
    else:
        print(f"❌ Extraction Failed: {result.get('error')} - {result.get('message')}")

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

{
  "success": true,
  "data": {
    "vin": "5UXWX7C51BA123456",
    "confidence": "High",
    "carrier_type": "windshield"
  }
}