License Plate Recognition — Data That Sees the Road Ahead

In every breakthrough, there’s a moment of clarity — a realization that the future doesn’t simply arrive. It’s built by those who dare to innovate.

The License Plate Recognition dataset is designed for those builders. The engineers, researchers, and visionaries who understand that artificial intelligence isn’t just about automation — it’s about creating smarter, safer, and more responsive systems that interact meaningfully with the real world.

Originally curated by Roboflow Universe and now hosted on Kaggle, this dataset is more than just a collection of labeled images. It’s a toolkit for those who are ready to shape the future of intelligent transportation.

What’s Inside

This dataset is crafted not just for scale, but for significance. It’s detailed, diverse, and immediately usable for practical applications.

• 10,125 real-world images of license plates
• High-quality bounding box annotations included for each image
• Supports export in multiple industry-standard formats (YOLO, COCO, Pascal VOC, etc.)
• Ready to integrate with TensorFlow, PyTorch, YOLOv5, YOLOv8, and other popular ML frameworks
• Contains images under varied lighting, weather, and environmental conditions, ensuring robust model performance in real-world scenarios

Designed for Real-World Innovation

This dataset enables developers to build and deploy:

• Automatic Number Plate Recognition (ANPR) systems
• Smart surveillance and traffic monitoring solutions
• Toll booth and parking automation systems
• Intelligent vehicle access and fleet management platforms

It’s not just about recognizing license plates. It’s about building the infrastructure of tomorrow — from smart cities to secure transit systems.

Why This Dataset Matters

In a world increasingly driven by data, not all datasets are created equal. What makes this one powerful is its balance of quality, realism, and flexibility.

• Captured from real-world environments, not synthetic simulations
• Includes diverse backgrounds, plate styles, and viewing angles
• Easily integrates with most computer vision pipelines
• Saves time with pre-annotated, ready-to-train data

Whether you’re prototyping a solution for embedded devices or scaling an enterprise-grade platform, this dataset provides a reliable and versatile foundation.

For Learners, Builders, and Trailblazers

Whether you're a student experimenting with computer vision or an ML engineer deploying production models, this dataset is a place to begin — or go further.

Use it to explore the power of object detection. To build solutions that matter. To take an idea from concept to deployment.

Because in the end, license plate recognition isn’t just about vehicles or images. It’s about helping machines understand the world — so we can make that world more intelligent, more efficient, and more human.

Licensed Plate - Character Recognition for LPR, ALPR and ANPR

The dataset features license plates from 32+ countries and includes 1,200,000+ images with OCR. It focuses on plate recognitions and related detection systems, providing detailed information on plate numbers, country, bbox labeling and other data as well as corresponding masks for recognition tasks - Get the data

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F22059654%2F40ac2244af1a25663454889e12e6e75a%2FFrame%20129.png?generation=1726156668935646&alt=media" alt="">

The dataset encompasses plate detection systems, cameras, and character recognition for accurate identification of license plates. LPR systems, including ALPR and ANPR, are utilized for automatic license and number plate detection, with models recognizing characters and identifying vehicles in real time. It supports object detection, recognition algorithms, and LPR cameras, ensuring high accuracy across different regions and environments.

Countries inthe dataset

Ukraine, Lithuania, Serbia, Turkey, Kazakhstan, Latvia, Belarus, Bahrain, Estonia, Uzbekistan, Moldova, Vietnam, Armenia, UAE, Georgia, Brazil, Finland, Azerbaijan, Kyrgyzstan, Egypt, Thailand, Mexico, Argentina, India, KSA, Pakistan, Morocco, Tajikistan, Mongolia, Palestine, Turkmenistan and other countries.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F22059654%2F698d1436d67cbe64ee2a9e4e8bbabdaf%2FFrame%20128.png?generation=1726156074455311&alt=media" alt="">

💵 Buy the Dataset: This is a limited preview of the data. To access the full dataset, please contact us at https://unidata.pro to discuss your requirements and pricing options.

Applications range from parking management to security systems, offering real-time data processing and vehicle registration tracking. This comprehensive dataset is ideal for advancing automated systems for plate readers, ALPR technology, and solutions for vehicle registration, security, and enforcement.

Variables in .csv files:

• file_name: filename of the vehicle photo
• license_plate.country: country where the vehicle was captured
• bbox: bounding box coordinates for the vehicle
• license_plate.visibility: visibility of the license plate
• license_plate.id: unique license plate identifier
• license_plate.mask: normalized coordinates of the license plate
• license_plate.rows_count: number of lines on the license plate
• license_plate.number: recognized text on the license plate
• license_plate.serial: series identifier for UAE plates
• license_plate.region: subregion for UAE plates
• license_plate.color: color of the plate code for Saudi Arabia

🌐 UniData provides high-quality datasets, content moderation, data collection and annotation for your AI/ML projects

License Plate Recognition - 89 986 Image

Dataset contains 89,986 photos of license plates (number plates), designed for OCR tasks, vehicle registration analysis, and traffic management systems. This large-scale collection provides real-world data for training and evaluating models in license plate recognition (LPR), character recognition, and autonomous driving applications. - Get the data

  Dataset characteristics:

Characteristic Data Description License plate… See the full description on the dataset page: https://huggingface.co/datasets/ud-smart-city/united-states-license-plate-dataset.

Dataset of license plate recognition

Dataset offers 89,986 images of vehicles featuring license plates from the USA, making it an excellent resource for tasks involving OCR (Optical Character Recognition), license plate identification, and vehicle registration data extraction. Each image is accompanied by a CSV file that provides the corresponding plate text and country code, ideal for developing and testing text recognition systems. With this dataset, researchers and developers can… See the full description on the dataset page: https://huggingface.co/datasets/UniDataPro/united-states-license-plate-dataset.

The Indonesian License Plate Dataset is an open-source collection of images containing various vehicle license plates from Indonesia. The dataset aims to facilitate research and development in license plate recognition, computer vision, and machine learning applications. It is suitable for training and evaluating license plate detection and recognition algorithms.

Source and Authors: The dataset is maintained and made available by two contributors: 1. Irvan Teady Sentosa - URL: Irvan Teady Sentosa - Indonesian License Plate Dataset - Published on Roboflow Universe, a platform for sharing and accessing datasets. - Visited on: 19th June 2023.

1. Habib Robbani
   • URL: Habib Robbani - Indonesian Motorcycle Plate Dataset
   • Published on Roboflow Universe.
   • Visited on: 19th June 2023.

Content: The dataset comprises a diverse collection of images capturing different types of vehicle license plates used in Indonesia. It includes various lighting conditions, weather scenarios, and plate orientations to ensure robustness in model training and testing. The images may also include motorbikes, as one of the datasets explicitly focuses on Indonesian motorcycle plates.

Use Cases: The Indonesian License Plate Dataset is valuable for various purposes, including but not limited to: - License plate detection and localization. - License plate character recognition. - Vehicle type classification (e.g., car, motorcycle). - Traffic analysis and surveillance systems. - Autonomous vehicle applications. - Research and experimentation in computer vision and machine learning.

Terms of Use: As the dataset is open-source, it is typically governed by a specific license, which might be provided on the Roboflow Universe platform or through the dataset's original sources. Users are encouraged to abide by the terms and conditions set forth by the dataset authors and platform.

Final Project in University: The Indonesian License Plate Dataset could be utilized as a fundamental resource for a final project at the university level. Students or researchers working on projects related to computer vision, deep learning, or image processing might find this dataset particularly useful for training models and evaluating their performance in the context of Indonesian license plate recognition and related applications.

License Plate Recognition - 2.6 Million images

Dataset comprises 2.6 million images of vehicle license plates across 86 countries, providing a comprehensive resource for OCR, traffic analysis, and autonomous vehicle systems.It focuses on plate recognitions and related detection systems, providing detailed information on plate numbers, country, bbox labeling and other data as well as corresponding masks for recognition tasks.- Get the data

  Dataset characteristics:… See the full description on the dataset page: https://huggingface.co/datasets/ud-smart-city/license-plate-dataset.

This dataset contains 10,125 high-quality images annotated for license plate detection using bounding boxes. It is designed for object detection tasks and is well-suited for building Automatic Number Plate Recognition (ANPR) systems. The dataset covers a wide range of vehicle types and imaging conditions, supporting robust model training and evaluation. Images are auto-oriented and split into training, validation, and test sets following a 70/20/10 ratio. The dataset is compatible with multiple popular formats, including YOLO (v5–v11) and COCO, making it ideal for real-time traffic monitoring, parking management, and intelligent transportation applications.

• This dataset contains 1695 images, each with bounding box annotations for license plate.
• The annotations are provided in the YOLO text format, which includes class labels and bounding box coordinates.
• This dataset is useful for object detection tasks such as license plate detection and recognition.

[Dataset Description] This dataset was constructed to facilitate the development of methods aimed at improving image quality and recognition accuracy of license plates captured from dashcam videos. It contains paired low-quality license plate video sequences and corresponding high-quality target images, enabling quality enhancement model training and evaluation. Each sample includes a sequence of 31 low-quality license plate frames alongside a single high-quality license plate image that serves as the target (pseudo-ground truth) for quality enhancement. This dataset can be used for training models to enhance image quality and includes recognition labels for assessing recognition accuracy. Additionally, the dataset provides outputs obtained using the MF-LPR² model proposed in our paper, offering a benchmark for performance comparison.

[Dataset Contents] 1) Low-Quality (LQ) License Plate Image Sequences
- 31 frames per sequence. 2) High-Quality (HQ) License Plate Images (Pseudo-Ground Truth)
- Includes both the original images and versions cropped to the Region of Interest (ROI). - ROI coordinates provided. 3) License Plate Recognition Label
- For privacy reasons, only the numerical characters of the license plate are provided. 4) MF-LPR² Model Super-Resolution (SR) Results
- Outputs generated by the proposed MF-LPR² model
5) Homography Transformation Coordinates
- Coordinates for aligning low- and high-quality images
- Utility tool is also provided

[Citation] When using this dataset, please cite the following paper: Na, K., Oh, J., Cho, Y., Kim, B., Cho, S., Choi, J., & Kim, I. (2025). MF-LPR2: Multi-frame license plate image restoration and recognition using optical flow. Computer Vision and Image Understanding, 256, 104361.

[Disclaimer]
All Korean Hangul characters in the license plate regions have been pseudonymized using targeted blurring techniques. In the low-resolution (LR) image sequences, the frames have been cropped to the license plate region to minimize the exposure of the surrounding vehicle or environment. In some cases, small portions of the vehicle may remain visible due to cropping margins. The recognition labels contain only numerical values. No GPS metadata, timestamps, vehicle make/model information, or environmental background data is included in this dataset.

This dataset is released strictly for non-commercial, academic research purposes. By accessing or using this dataset, you agree to the following terms: * You will not attempt any form of re-identification or misuse of the data. * You will not use the dataset for surveillance, law enforcement, or any commercial applications. * You will cite the original paper when publishing any results that use this dataset. * You will comply with all applicable data protection and privacy laws. The authors and their affiliated institutions assume no legal responsibility for any use of this dataset that extends beyond its intended academic research purpose.

USA License Plate Dataset containing labeled vehicle images from various states to support AI-driven law enforcement, parking automation, and access control systems

A meticulously curated dataset containing license plate images from multiple countries and regions, designed for training AI-based License Plate Recognition (LPR) and Intelligent Transportation Systems (ITS).

Explore the European License Plate Recognition Vital for AI scene analysis, content filtering, and smart user interfaces

License Plate Recognition Dataset

Dataset comprises 118,798 images of vehicle license plates, meticulously annotated for OCR tasks, object detection, and real-time traffic analysis. Designed to advance research in autonomous vehicles, traffic management, and urban mobility. By leveraging this dataset, researchers and developers can enhance license plate recognition (LPR) systems, improve traffic monitoring, and develop intelligent transportation solutions - Get the data The… See the full description on the dataset page: https://huggingface.co/datasets/UniDataPro/united-kingdom-license-plate-dataset.

Car License Plate Detection Dataset offers diverse vehicle images, aiding AI in accurately recognizing and interpreting license plates

License Plate Recognition - 58 229 Image

Dataset offers 58 229 images of vehicles featuring license plates from Belgium, making it an excellent resource for tasks involving OCR (Optical Character Recognition), license plate identification, and vehicle registration data extraction. Each image is accompanied by a CSV file that provides the corresponding plate text and country code, ideal for developing and testing text recognition systems. - Get the data

  Dataset… See the full description on the dataset page: https://huggingface.co/datasets/ud-smart-city/belgium-license-plate-dataset.

License Plate Recognition Dataset

Dataset contains 196 664 photos of license plates, providing a comprehensive resource for OCR (Optical Character Recognition), plate recognition, and vehicle registration analysis. Each image file is paired with a CSV file containing license plate text for use in OCR, plate recognition, and characters recognition tasks. By leveraging this dataset, researchers can explore traffic patterns, improve traffic management, and train learning models for… See the full description on the dataset page: https://huggingface.co/datasets/UniDataPro/poland-license-plate-dataset.

The Fifty States Car License Plates Dataset features diverse car images captured under varied conditions, each containing visible license plate numbers along with additional text elements. This dataset is designed to train and test machine learning models for license plate recognition, text differentiation, and computer vision tasks in real-world scenarios.

Dataset of license plate recognition

Dataset contains 177,827 photos of license plates, providing a comprehensive resource for OCR (Optical Character Recognition), plate recognition, and vehicle registration analysis. Each image file is paired with a CSV file containing license plate text for use in OCR, plate recognition, and characters recognition tasks. By leveraging this dataset, researchers can explore traffic patterns, improve traffic management, and train learning models for… See the full description on the dataset page: https://huggingface.co/datasets/UniDataPro/germany-license-plate-dataset.

License Plate Recognition

## Overview

License Plate Recognition is a dataset for object detection tasks - it contains License Plates annotations for 6,407 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

License Plate Recognition

## Overview

License Plate Recognition is a dataset for object detection tasks - it contains License Plates annotations for 1,235 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).