Please note: The algorithm descriptions in English have been automatically translated. Errors may have been introduced in this process. For the original descriptions, go to the Dutch version of the Algorithm Register.

Parking enforcement using a scanning vehicle

The council (acting on its behalf: ParkerenDelft B.V.) uses scanning vehicles to carry out parking checks. The scanning vehicle uses an image recognition algorithm to 1) recognise number plates and 2) check the parking rights database to see whether the vehicle is authorised to park there. Enforcement officers also have handheld scanners which they use to carry out checks. 
Last change on 14th of July 2026, at 14:35 (CET) | Publication Standard 1.0
Publication category
Impactful algorithms
Impact assessment
DPIA, AIIA
Status
In use

General information

Theme

Traffic

Begin date

2020-01

Contact information

privacy@delft.nl

Link to publication website

https://parkerendelft.com/ https://www.delft.nl/parkeren-delft

Responsible use

Goal and impact

The purpose of the scanning vehicle is to check parked vehicles in areas where paid parking applies or where a parking permit is required. The use of the scanning vehicle allows more cars to be checked efficiently. Members of the public come into contact with it when they park their vehicles in the zones where the scanning vehicle is operating.

The cameras, and consequently the algorithm, are used to enforce regulations in designated parking areas for vehicles. The algorithm has a minimal impact on members of the public. The algorithm merely reads the number plate. The images of the number plates of alleged offenders are manually checked by an authorised enforcement officer to determine whether an offence has actually been committed. If this is not the case, the system is corrected and the images are deleted within 72 hours.

Considerations

Car parking and traffic take up a great deal of the limited space available, and this affects the quality of life in the city.

The deployment of a scanning vehicle in the municipality of Delft is essential for more efficient and effective parking enforcement. This is because the scanning vehicle automatically scans number plates and checks for paid parking, thereby increasing the likelihood of parking fines for those who have not paid and improving compliance with parking regulations. In this way, parking spaces remain available for visitors and permit holders.

A key advantage of using the scanning vehicle and its accompanying algorithm is efficiency; fewer enforcement officers are needed to carry out checks, and checks are completed more quickly. Furthermore, the system promotes fairness, as the algorithm records and processes all number plates without distinction. One disadvantage is that the algorithm lacks the full context. For example, the cameras cannot always detect disabled parking permits, meaning that vehicles with such permits are wrongly flagged as being in breach of the rules. For this reason, human intervention is always required. An enforcement officer assesses the situation and issues a fine if necessary. This assessment can be carried out either remotely (via the image analysis module) or on-site (using a handheld device).

Human intervention

In the event of a possible offence, a human operator intervenes to carry out an additional check.

If it transpires that there is no valid parking authorisation, or if the system believes it has misread the number plate, the footage is sent to the enforcement officer (via their personal account). This person assesses the footage. In some cases, a decision may be made to send an enforcement officer/BOA with a handheld device to the location in question to check whether there are exceptional circumstances, or if the space is a disabled parking space.

If it transpires that:

a. a valid parking authorisation does exist after all, this is updated in the system and the photo is automatically deleted within 72 hours;

b. there is indeed no valid parking authorisation, this will be confirmed in the system.

If a supplementary assessment* is issued, the images are stored in Brickyard, where the offences are retained.

* (additional parking charge notice (nha)).

Risk management

There is no live video footage; only digital images (photographs) containing number plate information are used. Only potential offenders are temporarily placed on a list. This list is checked manually by an enforcement officer or BOA within 72 hours. Information about non-offenders is not stored.

In the event of a possible offence, a human intervention takes place to carry out an additional check.

Legal basis

The parking policy (paid parking and parking permits) regulates the parking behaviour of residents and visitors in designated areas of the city. In practice, there are two types:

1. Pay-and-display parking in designated spaces.

2. A permit issued and paid for, allowing parking in designated spaces.

This is laid down in Articles 225, 234 and 235 of the Local Government Act and the Municipal Parking Charges Order.

Links to legal bases

  • Local Government Act, Article 225: https://wetten.overheid.nl/jci1.3:c:BWBR0005416&titeldeel=IV&hoofdstuk=XV&paragraaf=3&artikel=225&z=2026-06-04&g=2026-06-04
  • Local Government Act, Article 234: https://wetten.overheid.nl/jci1.3:c:BWBR0005416&titeldeel=IV&hoofdstuk=XV&paragraaf=4&artikel=234&z=2026-06-04&g=2026-06-04
  • Local Government Act, Article 235: https://wetten.overheid.nl/jci1.3:c:BWBR0005416&titeldeel=IV&hoofdstuk=XV&paragraaf=4&artikel=235&z=2026-06-04&g=2026-06-04

Elaboration on impact assessments

The local authority carried out a DPIA and an AIA to assess the impact of the solution.

Impact assessment

  • Data Protection Impact Assessment (DPIA)
  • AI Impact Assessment (AIIA)

Operations

Data

The scanning vehicle collects all the data required to verify parking authorisation: location, number plates, time and photographs of the immediate surroundings of parked cars. When a scanning vehicle scans a number plate, the location, number plate and time are sent to the back office. The back office checks whether the vehicle has paid for parking or holds a permit. If so, the data in the back office and the photographs in the scanning vehicle are deleted immediately and are not sent to the back office. If this is not the case, the photos are retrieved from the scanning vehicle and sent to the back office. After a configurable period (72 hours), the data is automatically deleted if the enforcement officer has not taken any action (issue a fine, issue a warning or dismiss the case).

Links to data sources

  • RDW: https://www.rdw.nl/
  • NPR: https://www.nationaalparkeerregister.nl/

Technical design

The scanning vehicle drives through the streets on public roads in areas where paid parking has been introduced, or where a permit is required. During these journeys, various types of cameras on the scanning vehicle (infrared cameras, colour cameras and panoramic cameras) automatically scan the number plates of all parked cars within the enforcement zone. Each scanned number plate (photo) is then linked to an exact geographical position and parking time.

This data is securely transmitted to Brickyard’s back office. There, an algorithm checks whether the geographical position of the number plate is within a parking bay as defined on a map provided by the local authority. This map contains all relevant information about the parking zones. Following this check, the system filters the data and consults the parking entitlements in the National Parking Register.

If it transpires that no valid parking entitlement has been registered for the number plate in question (i.e. that there is a permit, a GPK or that paid parking has been activated), an enforcement officer uses the image assessment module to check whether there are any special circumstances. If necessary, a parking inspector may be sent to the vehicle to assess the situation on the spot. If there are no exceptional circumstances, the enforcement officer confirms the offence, after which a supplementary parking charge notice is issued.


External provider

Brickyard B.V.

Similar algorithm descriptions

  • Parking enforcement using a scanning vehicle (via ParkeerService)
    Last change on 2nd of July 2026, at 12:07 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    DPIA
    Status
    In use
  • The council uses automated number plate recognition vehicles to carry out parking checks. The cameras record number plates using an image recognition algorithm and check them against the parking permit database to verify that a valid parking permit is in place. In addition, special investigating officers carry out checks using handheld scanners.
    Last change on 17th of June 2026, at 9:20 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    DPIA
    Status
    In use
  • To carry out parking controls, the municipality uses scan cars. The cameras on the scan car use an image recognition algorithm to 1) recognise license plates and 2) check in the parking rights database whether the vehicle is allowed to park there. Boas also have manual scanners that perform checks.
    Last change on 1st of December 2025, at 9:53 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    DPIA
    Status
    In use
  • The municipality monitors the enforcement of paid parking of vehicles using a scanning car with a camera. The scan car drives on roads within the paid parking area; records license plates of vehicles in parking bays; and checks whether a parking right is present. After physical inspection by parking enforcement officers, it is determined whether a fine will be imposed.
    Last change on 9th of June 2026, at 11:29 (CET) | Publication Standard 1.0
    Publication category
    Other algorithms
    Impact assessment
    DPIA
    Status
    In use
  • The municipality conducts parking checks using scan cars. The cameras on the scan car use an image recognition algorithm to recognise license plates and check whether the vehicle is allowed to park there.
    Last change on 5th of January 2024, at 12:40 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use