marearts-anpr 3.8.3

MareArts ANPR (Automatic Number Plate Recognition) library

MareArts ANPR SDK

ANPR Detection	Road Objects Detection	Mobile App

Automatic Number Plate Recognition (ANPR) SDK for multiple regions with GPU acceleration support.

💎 One License, All Access: SDK + Mobile App + Road Objects Detection - Use everywhere with a single license.

---

🎉 MareArts ANPR Mobile App - v1.9.4

📱 Now available on iOS! Android coming soon.

Fast, accurate, on-device license plate recognition for parking management, security, and vehicle tracking.

✨ Key Features:

• 🚀 Fast on-device AI processing (~100-160ms)
• 🔗 Webhook integration - Send detections to Discord, Slack, or your own server
• 📊 CSV export with bounding box, notes, reporter, auto-detected device info
• 🔄 Cloud sync - Two-way background sync across devices
• 📋 Smart rules grouping - Organized A-Z for easy navigation
• 🌐 Download rules from web - Upload on marearts.com, download on phone
• 🔒 100% offline capable - privacy first
• 🗺️ Map view with GPS tracking
• ✅ Whitelist/Blacklist management
• 🌍 Multi-region support (Universal, Europe+, Korea, North America, China)

📖 Read the complete Mobile App Guide →

Use the mobile app as your ANPR license - no additional purchase required.
Get your license at marearts.com/products/anpr

---

🌐 NEW: ANPR Management Server

Professional server with REST API and Web Dashboard

Deploy a complete ANPR management system with real-time monitoring, detection history, and visual analytics.

Quick Start:

cd management_server
pip install -r requirements.txt  # First time only
ma-anpr config                   # Configure credentials
python server.py                 # Start server
# Open http://localhost:8000/

Features: REST API, Web Dashboard, Real-time logs, SQLite database, Live model switching

📖 Full Documentation →

---

🎉 MareArts SDK v3.8.x

V15 OCR - Next Generation Recognition ⭐

• 🎯 Improved Accuracy: Enhanced recognition across all regions
• 📝 Better Multi-line Handling: Improved recognition of plates with multiple text lines
• 🚀 Better Performance: Higher accuracy with strong real-time throughput
• 🔄 Easy Upgrade to V15: Simple drop-in replacement for V14 OCR
• ✅ Recommended: V15 OCR is now the default for new projects

Backward Compatible: V14 OCR continues to be fully supported

---

MareArts ANPR SDK Features

• 🌍 Multi-Region Support: Korean, Europe+, North America, China, and Universal license plates
• 🔄 Dynamic Region Switching: Change regions instantly with set_region() without model reload
• ⚡ GPU Acceleration: CUDA, DirectML support for real-time processing
• 🎯 High Accuracy: Advanced models with regional vocabulary optimization
• 📦 Batch Processing: Process multiple plates simultaneously
• 🐳 Production Ready: Docker API with smart model caching and multi-architecture support

Quick Start

Installation

# CPU Installation
pip install marearts-anpr

# GPU Installation (CUDA, DirectML)
pip install marearts-anpr[gpu]        # NVIDIA CUDA
pip install marearts-anpr[directml]   # Windows GPU (AMD/Intel/NVIDIA)

📦 See complete installation guide

Basic Usage

from marearts_anpr import ma_anpr_detector_v14, ma_anpr_ocr_v15, marearts_anpr_from_image_file

# Initialize V14 Detector
detector = ma_anpr_detector_v14(
    "micro_320p_fp32",
    # 320p models (Fast): pico_320p_fp32/fp16, micro_320p_fp32/fp16, small_320p_fp32/fp16, medium_320p_fp32/fp16, large_320p_fp32/fp16
    # 640p models (High detection): pico_640p_fp32/fp16, micro_640p_fp32/fp16, small_640p_fp32/fp16, medium_640p_fp32/fp16, large_640p_fp32/fp16
    user_name,
    serial_key,
    signature,
    backend="cuda",  # cpu, cuda, directml (auto-selected if "auto")
    conf_thres=0.25,  # Detection confidence threshold (default: 0.25)
    iou_thres=0.5     # IoU threshold for NMS (default: 0.5)
)

# Initialize V15 OCR with regional vocabulary (Recommended - Latest)
ocr = ma_anpr_ocr_v15(
    "small_fp32",       # Model: pico_fp32, micro_fp32, small_fp32, medium_fp32, large_fp32
                        # int8 models available: pico_int8, micro_int8, small_int8, medium_int8, large_int8 (smaller, faster)
    "univ",             # Region: kor/kr, euplus/eup, na, china/cn, univ (choose specific region for best accuracy)
    user_name,
    serial_key,
    signature,
    backend="cuda",  # cpu, cuda, directml (auto-selected if "auto") 
)

# Or use V14 OCR (backward compatible)
# from marearts_anpr import ma_anpr_ocr_v14
# ocr = ma_anpr_ocr_v14("small_fp32", "univ", user_name, serial_key, signature, backend="cuda")

# Or use unified interface with version parameter
# from marearts_anpr import ma_anpr_ocr
# ocr = ma_anpr_ocr("small_fp32", "univ", user_name, serial_key, signature, version='v15', backend="cuda")  # v15: Latest, or version='v14': Stable  

# Process image
result = marearts_anpr_from_image_file(detector, ocr, "image.jpg")
print(result)
# Output: {'results': [{'ocr': 'ABC123', 'ocr_conf': 99, ...}], ...}

💡 🔄 Learn more about usage

Dynamic Region Switching

Switch regions without reinitialization (works with both V14 and V15 OCR):

ocr.set_region('euplus')  # Europe+ (or 'eup')
ocr.set_region('kr')   # Korean
ocr.set_region('na')   # North America
ocr.set_region('cn')   # China
ocr.set_region('univ') # Universal (all regions)

🔄 Learn more about dynamic region switching

Multiple Input Formats & CLI

From different image sources:

import cv2
from PIL import Image
from marearts_anpr import marearts_anpr_from_cv2, marearts_anpr_from_pil

result = marearts_anpr_from_cv2(detector, ocr, cv2.imread("image.jpg"))
result = marearts_anpr_from_pil(detector, ocr, Image.open("image.jpg"))
result = marearts_anpr_from_image_file(detector, ocr, "image.jpg")

CLI commands:

ma-anpr image.jpg                    # Process image (V15 OCR is default)
ma-anpr image.jpg --ocr-version v15  # Use V15 OCR (explicit)
ma-anpr image.jpg --ocr-version v14  # Use V14 OCR
ma-anpr test-api image.jpg           # Test API (1000/day limit)
ma-anpr validate                     # Validate license
ma-anpr models                       # List available V14 and V15 models

🔧 See complete usage examples and CLI reference

Model Performance

Detector Performance (V14)

Model Name	Detection Rate	Speed (GPU)	Notes
pico_320p_fp32	96.02%	129 FPS (7.8ms)	📱 Smallest + fast
pico_640p_fp32	98.54%	66 FPS (15.2ms)	Balanced
micro_320p_fp32	97.13%	128 FPS (7.8ms)	🏆 Best overall
micro_320p_fp16	97.13%	56 FPS (17.9ms)	🏆 Best mobile (50% smaller)
micro_640p_fp32	98.99%	68 FPS (14.6ms)	Highest detection
small_320p_fp32	98.00%	142 FPS (7.0ms)	⚡ Fastest
medium_320p_fp32	98.06%	136 FPS (7.4ms)	High detection
large_320p_fp32	98.40%	131 FPS (7.6ms)	Strong performance

Note: 320p models are 2× faster than 640p. FP16 models are 50% smaller with same detection rate.

OCR Performance (V15)

Average across all regions

Model Name	Exact Match	Character Accuracy	Speed (GPU)	Notes
pico_fp32	98.66%	99.74%	235.1 FPS (4.32ms)	📱 Smallest, fast
micro_fp32	99.01%	99.80%	245.8 FPS (4.07ms)	Fast with high accuracy
small_fp32	98.66%	99.75%	280.2 FPS (3.57ms)	⚡ Fastest inference
medium_fp32	99.13%	99.83%	254.5 FPS (3.92ms)	🎯 Best accuracy
large_fp32	98.99%	99.81%	241.5 FPS (4.14ms)	High accuracy

int8 Models (smaller files):

• pico_int8, micro_int8, small_int8, medium_int8, large_int8
• 75% smaller file size, similar accuracy

Supported Regions: Korea (kor or kr), Europe+ (euplus or eup), North America (na), China (china or cn), Universal (univ)

Note: Both short codes and full names are accepted

📊 See all models and benchmarks

Regional Support

MareArts ANPR supports license plates from multiple regions with specialized vocabulary optimization:

• 🇰🇷 Korean (kr) - Korean license plates with Hangul characters (best accuracy: 99.56%)
• 🇪🇺 Europe+ (eup) - EU countries + Albania, Andorra, Bosnia & Herzegovina, Indonesia, and more
• 🇺🇸🇨🇦🇲🇽 North America (na) - USA, Canada, and Mexico license plates
• 🇨🇳 China (cn) - Chinese license plates with province codes
• 🌍 Universal (univ) - All regions (default, but choose specific region for best accuracy)

💡 Dynamic Region Switching: Use ocr.set_region('kr') to switch regions instantly without reloading the model, saving ~180 MB per additional region.

🌍 See complete regional support and character sets

Documentation

• 📦 Installation Guide - Detailed installation options and requirements
• 🔧 Usage Examples - Python SDK, CLI usage, dynamic region switching, and environment variables
• 💻 Example Code - Basic, advanced, and batch processing examples
• 🚀 Model Versions - Available models, benchmarks, and performance metrics
• 🌍 Regional Support - Supported countries and character sets
• 🐳 Docker Deployment - Container setup, API server, and multi-architecture builds
• 🧪 Try ANPR - Test our ANPR without license (1000 requests/day)
• ❓ FAQ - Licensing, regions, features, and troubleshooting

MareArts Ecosystem

Explore our AI toolkit:

• marearts-anpr - Automatic Number Plate Recognition (GitHub)
• 🎉 marearts-anpr Mobile App - ANPR on iOS & Android (App Store | Guide)
• marearts-road-objects - Road object detection for persons, vehicles, and 2-wheelers (GitHub)
• marearts-xcolor - Color extraction and similarity analysis (GitHub)
• marearts-mast - Real-time panoramic stitching (GitHub)
• marearts-crystal - Encryption and decryption toolkit (PyPI)

Support & Resources

Resource	Link
📧 Contact	[e-mail]
🏠 Homepage	https://marearts.com
💳 License Purchase	ANPR Solution
🎮 Live Demo	http://live.marearts.com
📺 Video Examples	YouTube Playlist

License

© 2024 MareArts. All rights reserved.

This software requires a valid license key. Visit MareArts ANPR Solution for licensing options.

---