License Plate Recognition Dataset

Dataset contains 88,753 high-resolution images of vehicle license plates captured across diverse real-world conditions in the Netherlands. Designed to advance research and development in automatic license plate recognition (ALPR), optical character recognition (OCR), and intelligent traffic management systems. With this dataset, researchers and developers can build and refine solutions for traffic analysis, transportation management, and autonomous… See the full description on the dataset page: https://huggingface.co/datasets/UniDataPro/netherlands-license-plate-dataset.

Netherlands License Plate Dataset with annotated images to power computer vision models for automated number plate recognition, security, and smart parking

neotje/dutch-licence-plate-ANPR dataset hosted on Hugging Face and contributed by the HF Datasets community

Dutch License Plate Recognition

## Overview

Dutch License Plate Recognition is a dataset for object detection tasks - it contains Dutch LP Recognition annotations for 530 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).

Context

This is a dataset with images of (mostly Dutch) license plates of all kinds of vehicles. With this data you could create a machine learning model that detects and locates license plates.

Here are a few use cases for this project:

1. Traffic Monitoring: Dutch-Lpr can be employed by Dutch law enforcement agencies to monitor vehicle license plates at various checkpoints, toll booths, and traffic junctions, aiding in tracking traffic violations, such as speeding or running red lights, and improving overall road safety.

2. Stolen Vehicle Detection: Authorities can utilize Dutch-Lpr to cross-check license plates against a database of reported stolen vehicles in real-time, allowing for faster identification and recovery of these vehicles.

3. Parking Management: Dutch-Lpr can be integrated into smart parking systems to automatically recognize license plates of vehicles entering and leaving parking lots or garages. This enables more efficient management and billing of parking spaces, as well as tracking of vehicle occupancy within the lot.

4. Access Control: Dutch-Lpr can be used to automate access control in gated communities, private properties, or restricted-access facilities by granting entry only to authorized vehicles with pre-registered license plates, enhancing security and reducing the need for manual identification.

5. Road Usage and Congestion Analysis: Dutch-Lpr can facilitate the study of road usage patterns by analyzing the license plates of vehicles passing through designated areas. This data can contribute to the planning and implementation of improved transportation infrastructure, traffic management policies, and congestion reduction initiatives.

Traffic performance of lorries and tractors with a Dutch registration number, expressed in vehicle kilometres per year, broken down by vehicle characteristics and by territory. From the register of the Stichting Nationale Autopas (NAP), the license plates of the total sample population have been enriched with mileage readings and corresponding registration dates. Based on the differences between odometer readings of two consecutive registration dates, the kilometres travelled for the period between the registration dates were calculated. This means that annual mileages have been calculated for all vehicles. These annual mileages are used to estimate, among other things, the traffic performance of trucks and tractors.

More information on how to access the data:

https://www.cbs.nl/en-en/our-services/custom-and-microdata/microdata-self-research

Methodology

Secondary observation using the registers of the Road Traffic Administration (RDW)

Population

All lorries and tractors registered in the Netherlands with a valid Dutch registration number that have been admitted to traffic on public roads during (part of) the reporting year, including lorries and tractors that belong to the company stock in (part of) the year and lorries and tractors that are only active on the road network for part of the year, mainly new or imported vehicles and vehicles that have been scrapped or exported during the year. As of 2018, vehicles that have not been insured in the entire previous year are excluded.

This find is registered at Portable Antiquities of the Netherlands with number PAN-00094589

This table contains figures on average annual mileage of passenger cars with Dutch license plates by type of ownership, fuel type, weight class and age class of the vehicle. The figures for the 2020 reporting year have been corrected for the smoothing effect of the method by means of a correction factor. This smoothing effect smoothes out the annual variation in the figures. This gives a distorted picture of periods in which mobility suddenly changes radically, such as in 2020 as a result of the corona crisis. The vehicle population for which mileage is estimated is based on motor vehicle fleet statistics. The population of the figures in this table is based on the new motor vehicle fleet selection method. The difference between the old and the new selection method is described in a method report, see section 4. The series of kilometers estimated on the basis of the new population is available from reporting year 2018. The series based on the old vehicle population runs until and with reporting year 2020. The way in which the kilometers are estimated has not changed, only the population. Data available from: 2018 Status of the figures: The figures up to and including 2020 are final and the figures for 2021 have a provisional status. Changes as of November 11, 2022: None, this is a new table. When will new numbers come out? The new figures are expected at the end of 2023.

This table contains data on the ownership of mopeds, light mopeds and microcars as at 1 January of each year. The table excludes all other motor vehicles with a valid moped license plate, such as moped-quads, cargo mopeds and 3-wheeled mopeds. The data is based on information from the vehicle registration register of the Road Traffic Department (RDW). Since September 1, 2005 registration of registration for these vehicles is mandatory. The technical data of the vehicle and the personal details of the owner/holder are registered for all Dutch mopeds and light mopeds. The counts are subdivided by characteristics of the vehicle (type) and by characteristics of the owner (municipality). Annual figures are available from 2007. Status of the figures: final Changes as of April 20, 2012: Publication of data in this table has been discontinued due to the following reconsiderations: Data on moped ownership are calculated and published in different ways. Based on data from the vehicle registration register of the RDW, figures are known about the numbers of mopeds, broken down by region and calculated in numbers per 1,000 of the population. This information has been published in the Mopeds table since April 2012; number per 1000 inhabitants, type of vehicle and region on 1 January (see the hyperlink in paragraph 3). The Netherlands Mobility Survey (MON) publishes about vehicle ownership (including moped ownership) among individuals and households and calculates this indicator on the basis of sample observations. Data on this can be found in the table Vehicle ownership by gender (see the hyperlink in section 3). When will new numbers come out? This table has been discontinued.

This find is registered at Portable Antiquities of the Netherlands with number PAN-00094612

This find is registered at Portable Antiquities of the Netherlands with number PAN-00132316