The Premier Python SDK for Marine HIN Extraction

The GPU Bottleneck in Marine Computer Vision

For Python data engineers and computer vision developers, processing boat hull images locally is a massive resource drain. Open-source libraries like Tesseract or EasyOCR struggle significantly with marine environments. High-glare fiberglass, salt-corroded metal plates, and varied stamping depths cause local models to output garbage data. Attempting to train a custom YOLO/CRNN model for Hull Identification Numbers requires thousands of annotated marine images and expensive GPU compute time.

The StructOCR Python Advantage

The StructOCR Python SDK offers a plug-and-play alternative for your data pipelines. Instead of maintaining local AI models, our marine HIN OCR API handles the perspective deskewing and glare reduction in the cloud. You simply pass an image path or bytes object, and our API returns a fully parsed Python dictionary. This allows you to scale marine data prefill for thousands of vessels per hour with zero local GPU overhead.

Ideal for Python Workflows

• Marine Data ETL Pipelines: Automate the extraction of hull data from large batches of marine surveyor photos, piping the results directly into Pandas DataFrames.
• Automated Valuation Models (AVM): Feed verified vessel manufacturer and year data into Machine Learning pricing models for the used boat market.
• Maritime OSINT Analytics: Analyze and catalog harbor footage by systematically extracting HINs from scraped marine listings or port cameras.

Implementation: Python SDK Usage

Install the SDK via `pip install structocr`. This script demonstrates how to extract and navigate the nested HIN dictionary.

Prerequisite: Python 3.7+ and `pip install structocr`

from structocr import StructOCR
import json

# 💰 Save 30%+ vs competitors. Get 200 free credits instantly:
# 👉 https://structocr.com/register

def process_marine_hin():
    # Initialize the client with your secret API Key
    client = StructOCR("YOUR_API_KEY_HERE")
    image_path = "./dataset/raw_boat_hulls/vessel_01.jpg"

    try:
        print(f"Analyzing marine image: {image_path}...")
        
        # The SDK automatically handles file I/O and Base64 encoding
        result = client.scan_hin(image_path)

        # Verify mathematical correctness via the is_valid flag
        if result.get('is_valid'):
            print("✅ HIN Successfully Extracted and Validated!")
            print(f"Raw HIN:    {result.get('hin_number')}")
            print(f"Confidence: {result.get('confidence')}\n")
            
            # Drill down into the deeply parsed data
            parsed_data = result.get('parsed', {})
            print("--- Extracted Attributes ---")
            print(f"Manufacturer Code: {parsed_data.get('manufacturer_code')}")
            print(f"Production Month:  {parsed_data.get('production_month')}")
            print(f"Model Year:        {parsed_data.get('model_year')}")
            
        else:
            # Handle invalid HIN formats (e.g., image was a random object)
            error_msg = result.get('validation_error', 'No recognizable HIN found.')
            print(f"❌ Validation Failed: {error_msg}")

    except Exception as e:
        print(f"SDK or Network Exception: {e}")

if __name__ == "__main__":
    process_marine_hin()

Sample JSON Dictionary Response

The SDK returns a native Python dictionary containing the exact breakdown of the USCG or ISO 10087 standard HIN.

{
  "hin_number": "US-YAMC0323F313",
  "is_valid": true,
  "validation_error": null,
  "confidence": "High",
  "parsed": {
    "country_code": "US",
    "manufacturer_code": "YAM",
    "serial_number": "C0323",
    "production_month": "June",
    "production_year_short": "3",
    "model_year": "2013"
  }
}