Automated Crane Tally: Container OCR API

The Challenge of Top-Down Tallying

Tallying containers as they are lifted by Ship-to-Shore (STS) or Rubber-Tired Gantry (RTG) cranes introduces extreme visual complexities. Cameras mounted high on spreader beams capture images at steep, shifting angles. As containers sway and rotate in mid-air, the ISO 6346 text often appears completely upside-down. Traditional OCR systems fail under these dynamic conditions, forcing ports to rely on slow, manual spotters. To see how our engine cuts through these baseline visual challenges, you can upload a test image to explore our core Container OCR capabilities completely free.

Rotation-Invariant OCR API

Delivered strictly as a lightweight REST API to ensure seamless integration into any terminal architecture without heavy SDK dependencies, this endpoint is powered by our Logistics Container OCR engine. It features complete rotation invariance—automatically reading 180-degree inverted text from top-down cranes without pre-processing. The engine applies strict ISO 6346 validation on the fly, instantly calculating check digits to guarantee perfect data integrity before it reaches your TOS.

Crane Automation Scenarios

Integration & AI Prompts

Call the endpoint directly via HTTP. To maintain maximum security and reduce latency at edge nodes, the API flexibly accepts direct multipart file uploads or Base64-encoded strings, avoiding the need to expose live crane camera feeds via public URLs.

import requests
import base64

# Prepare Base64 Image from local crane spreader capture
with open("crane_sts_4_capture.jpg", "rb") as image_file:
    base64_image = base64.b64encode(image_file.read()).decode('utf-8')

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

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

    if result.get('success'):
        data = result['data']
        print("✅ Container Code Extracted!")
        print(f"Full Number:  {data.get('container_number')}")
        print(f"Valid ISO:    {data.get('is_valid')}")
        print(f"Confidence:   {data.get('confidence')}")
    else:
        print(f"❌ Extraction Failed: {result.get('error')} - {result.get('message')}")

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

{
  "success": true,
  "data": {
    "container_number": "ZCSU8921655",
    "is_valid": true,
    "confidence": "High",
    "parsed": {
      "owner_code": "ZCS",
      "category": "U",
      "serial_number": "892165",
      "check_digit": "5"
    }
  }
}