Introduction

degirum-vehicle is a high-performance Python library for License Plate Recognition (LPR) and vehicle analytics in images and video. Built for production deployments with minimal code and support for CPU and edge AI accelerators. Code examples and usage tutorials are available in the DeGirum Vehicle Analytics repo.

Fast and efficient vehicle analytics with:

• License plate detection, text recognition, and tracking for images, video files, and live streams

• Multi-hardware support: CPU, GPU, and edge AI accelerators - see supported hardware

• Simple APIs: Minimal code to detect, track, and recognize license plates with easy batch processing

• Flexible configuration: Python or YAML config for models, thresholds, and tracking parameters

• Production-ready tracking: Real-time vehicle tracking, Bayesian text aggregation, and event notifications

• Robust recognition: Multi-frame text aggregation for improved accuracy

Licensing

degirum-vehicle is one of the application packages licensed by DeGirum. Licensing is managed through DeGirum AI Hub. Users need to create an AI Hub account and set up a workspace with the appropriate permissions to generate licenses for degirum-vehicle.

For complete information on application package licensing, see the Application Package Licensing Guide. For workspace plan details and pricing information, visit the Workspace Plans page and the DeGirum Pricing Page.

The library provides code and pipelines for license plate recognition workflows. Model licensing is separate from the library licensing:

• License plate detection and recognition models: Trained by DeGirum and can be used commercially when users license the degirum-vehicle package

For complete licensing details, see the Models Reference.

Getting Started

Start with Installation & Setup and Basic Concepts, then explore the Guides for your use case.