RoboVAI OCR Enterprise Core 🚗🆔👤

RoboVAI OCR Enterprise Core (robovai-ocr v1.0.5) is a production-grade Computer Vision & OCR system developed by Mohamed Shaban (محمد شعبان العتماني) — Founder & AI Architect at RoboVAI.

It provides specialized AI engines tailored for Smart Parking Management Systems, Barrier Gate Controls, and Identity Verification Systems:

1. ALPREngine: Automatic License Plate Recognition for Egyptian, Arabic & International plates.
2. IDEngine: Egyptian National ID & Driver License Deep Learning Engine (14-digit National ID decoding, full name, address, DOB, governorate, photo crop).
3. FaceEngine: Facial recognition, driver authorization matching & registration database.

📊 Benchmark Performance Results (2,085 Dataset Images)

Benchmark evaluated on full enterprise dataset (robovai_plates 2,085 images):

🌟 Key Architectural Features

• Multi-Scale TTA Inference: Automatically applies 1.5x upscaling on low-resolution or distant frames with adaptive fallback confidence (conf = 0.15).
• Split-Region Whitelisted OCR: Separates Egyptian plates into Left Region (Digits 0-9) and Right Region (Arabic Letters) to eliminate letter-digit confusion.
• National ID Decoder: Automatically decodes 14-digit Egyptian National IDs into Date of Birth, Gender, Century, and Governorate of origin.
• High-Throughput Async Pipeline: Decouples fast YOLO object tracking from heavier OCR background execution to maintain real-time RTSP/Webcam stream processing.
• Green Box Visualizer: Renders high-visibility green bounding box (0, 255, 0) with text label banners for live video feeds and Web Dashboards.

🚀 Quick Start & Installation

Option 1: Install from PyPI

pip install --upgrade robovai-ocr

Option 2: Install from Source (Development Mode)

git clone https://github.com/m0shaban/robovai_ocr_system.git
cd robovai_ocr_system
pip install -e .

💻 Python SDK Usage Example

from robovai_ocr import RoboVAIEngine, Config

# 1. Initialize RoboVAI Engine
engine = RoboVAIEngine(device="cpu") # Or "cuda:0" for GPU

# 2. Automatic License Plate Recognition (ALPR)
plate_result = engine.read_license_plate("car_image.jpg")
print(plate_result)
# Output:
# {
#   "status": "success",
#   "plate_number": "س ط أ 1 2 3",
#   "confidence": 0.96,
#   "bbox": [120, 300, 450, 420]
# }

# 3. Egyptian National ID & Driver License Extraction
id_result = engine.extract_national_id("id_card.jpg")
print(id_result)
# Output:
# {
#   "national_id": "29901010100123",
#   "full_name": "أحمد محمد علي",
#   "dob": "1999-01-01",
#   "gender": "Male",
#   "governorate": "Cairo"
# }

# 4. Driver Face Verification
face_result = engine.verify_driver_face("driver.jpg", db_path="data/driver_faces")
print(face_result)

📡 REST API & Web Dashboard Server

Launch the FastAPI Server & Control Center Dashboard on Port 8500:

# Launch server via CLI
robovai-server --host 0.0.0.0 --port 8500

• Interactive Control Center Dashboard: http://localhost:8500/
• OpenAPI / Swagger Documentation: http://localhost:8500/docs

Key REST & WebSocket Endpoints

• POST /api/v1/alpr: License plate detection & OCR extraction.
• POST /api/v1/id-card: Egyptian National ID & Driver License OCR.
• POST /api/v1/face: Driver facial recognition & database matching.
• POST /api/v1/pipeline: Asynchronous background queue processing.
• WS /ws/stream: Real-time WebSocket RTSP stream processing.

💻 Command Line Interface (CLI)

# Process single license plate
robovai-cli alpr --image path/to/car.jpg

# Process Egyptian National ID card
robovai-cli idcard --image path/to/id.jpg

👨‍💻 Author & Enterprise Contact

• Author: Mohamed Shaban (محمد شعبان العتماني)
• Role: AI Architect & Founder of RoboVAI
• Personal Website: https://msalatmani.org
• Company Website: https://robovai.tech
• PyPI Package: https://pypi.org/project/robovai-ocr/
• License: MIT License