Direct naar content

You are here:

  1. Home
  2. Algorithms

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.

Fireworks Explorer

Netherlands Forensic Institute

Investigation after a fireworks accident is difficult. Fireworks are blown apart in many pieces and shreds. Often far from the scene of the accident. For proper investigation, officers need to know what fireworks were used. They also need to know which pieces to look for. The app helps police do this. With pictures of scraps, the app can see which fireworks exploded.

Last change on 5th of December 2025, at 7:26 (CET) | Publication Standard 1.0
Publication category
Other algorithms
Impact assessment
Field not filled in.
Status
In use

General information

Theme

Public Order and Safety

Begin date

032024

Contact information

Als er vragen zijn over de toepassing van dit algoritme of over algoritmes in het algemeen dan adviseren we u contact op te nemen met de communicatieafdeling van het NFI. Daarnaast kunt u voor informatie over dit onderwerp terecht op https://nfi.nl/

Link to publication website

https://www.vuurwerkverkenner.nl/

Link to source registration

https://www.forensischinstituut.nl/actueel/nieuws/2024/05/17/ai-model-getraind-om-vuurwerk-sneller-te-herkennen-na-ontploffing

Responsible use

Goal and impact

Provide indication of the type of fireworks found

Considerations

The application helps police instantly recognise which (professional) fireworks have exploded, based on photos of snippets. The web application also immediately gives related relevant information.

Human intervention

People need to understand and control the results of the algorithm

Risk management

The software is open source and therefore available to everyone

Legal basis

Penal Code

Operations

Data

The NFI officers trained the AI model by showing it examples of fireworks, both intact labels and fragments. In collaboration with the Defence Explosive Ordnance Disposal Service (EOD), hundreds of pieces of heavy fireworks were blown up for this purpose. Thousands of scraps were then collected and photographed at high quality. The fireworks for the investigation were made available by the Office of the Public Prosecutor's Office and came from seizures.

Technical design

Input are images, processing is done with a Vision Transformer and gives as output an embedding. This embedding is compared with the embeddings of the labels in the database and, based on the similarity, the results are sorted

External provider

The Netherlands Forensic Institute

Link to code base

https://github.com/NetherlandsForensicInstitute/vuurwerkverkenner

Similar algorithm descriptions

  • Fireworks dashboard

    Municipality of Haarlemmermeer

    For the deployment on fireworks nuisance, we use a "fireworks dashboard" with locations, days and times. With this, a fireworks analysis is made to identify hot spot locations and deploy the 'boas (special investigating officer) more efficiently.

    Last change on 7th of August 2024, at 9:53 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use

  • Supervision and enforcement of fireworks

    Municipality of Utrecht

    By being able to enforce fireworks nuisance in a better and more targeted way, we can properly enforce the General Local Bye-Law (APV).

    Last change on 5th of September 2025, at 13:03 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Uthiek, DPIA
    Status
    In use

  • Detecting risks of inaccuracies in customs declarations for Fireworks

    Customs

    This algorithm helps Customs to select clients for controls based on risk. Among other things, it uses declaration data from companies and assesses whether or not there are risks in bringing fireworks consignments into the European Union.

    Last change on 2nd of April 2025, at 12:44 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use

  • Road inspection - Kiwa KOAC

    Province of Groningen

    Inspection of provincial roads using LiDAR or laser technology. Using these techniques, the road section can be scanned in great detail and all unevenness highlighted. From this, the algorithm selects damage images on the basis of which the province will see what maintenance is needed here.

    Last change on 5th of February 2025, at 9:11 (CET) | Publication Standard 1.0
    Publication category
    Other algorithms
    Impact assessment
    Field not filled in.
    Status
    In use

  • Crack detection flood defences

    Hoogheemraadschap Hollands Noorderkwartier

    Detecting cracks in embankments using AI

    Last change on 30th of October 2024, at 7:41 (CET) | Publication Standard 1.0
    Publication category
    Other algorithms
    Impact assessment
    Field not filled in.
    Status
    In development