Direct naar content

You are here:

  1. Home
  2. Algorithms
  3. Automatic face comparison

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.

Automatic face comparison

Netherlands Forensic Institute

In an expert examination, facial images are compared. The facial image comparison aims to determine whether a person visible in camera images (crime suspect) and the image of a known face (police photo of a suspect) are of the same person or two different people.

Last change on 25th of June 2024, at 16:15 (CET) | Publication Standard 1.0
Publication category
High-Risk AI-system
Impact assessment
Field not filled in.
Status
In use

General information

Theme

Public Order and Safety

Begin date

-

Contact information

Als er vragen zijn over de toepassing van dit algoritme, of over algoritmen 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 www.nfi.nl

Responsible use

Goal and impact

The algorithm automatic facial image comparison supports the decisions made by experts. It is deployed in addition to the research done only by the experts. The algorithm is deployed to make the final judgment even more objective because of an algorithm can be well measured in what way and how well it works.

Considerations

The use of the algorithm allows the facial image comparison investigation to be carried out more efficiently and effectively. Combined with the assessment of the images by a human, the result of the investigation is more reliable and has more probative value in a criminal investigation.

Human intervention

The face image comparison research in which the algorithm is used has human intervention. It is determined beforehand whether the algorithm is suitable for the research material; afterwards, it is determined exactly what the outcome means for the research question. The result of the automatic face image comparison is included in the outcome of the research where a human watches and compares the face images. The algorithm is used alongside the expert's judgement.

Risk management

specification follows

Operations

Data

The algorithm uses a face image of a person on a camera image (often an unknown person) and a face image of a known person (often a police photo of a suspect). The facial images are personal data.

Technical design

The disputed face image and the reference image are presented to the software. The software uses an algorithm that extracts features of the face from each of the images and determines the degree of similarity between the facial images. The degree of similarity between the two images is expressed as a score. The interpretation of the score is again done by experts. For each case, we reassess whether the algorithm can be used.

External provider

The algorithm was developed by the Netherlands Forensic Institute.

Similar algorithm descriptions

  • Automated face comparison

    Municipality of Gooise Meren

    Automated face comparison

    Last change on 20th of June 2024, at 9:43 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use

  • Camera identification

    Netherlands Forensic Institute

    An investigation by an expert examines whether certain digital images (photos and/or videos) were taken with a certain digital camera and/or mobile phone. To answer this question, the camera identification algorithm is deployed, which identifies the camera.

    Last change on 25th of June 2024, at 16:19 (CET) | Publication Standard 1.0
    Publication category
    High-Risk AI-system
    Impact assessment
    Field not filled in.
    Status
    In use

  • Camera surveillance traffic measure (closed)

    Municipality of Haarlemmermeer

    Automatic Number Plate Recognition cameras (ANPR) at four different locations record license plates of passing motor vehicles. The algorithm checks whether an exemption has been granted and determines whether photos are submitted for review in the enforcement system.

    Last change on 3rd of September 2024, at 6:04 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use

  • Pressure measurement - blurring of the image

    Municipality of The Hague

    Algorithm that on the Scheveningen promenade ensures that (after counting and determining group dynamics) the camera clip is blurred.

    Last change on 5th of July 2024, at 8:04 (CET) | Publication Standard 1.0
    Publication category
    Other algorithms
    Impact assessment
    DPIA
    Status
    Out of use

  • Deployment of automated face comparison

    Municipality of Arnhem

    Automated facial comparison for first registration in the Basic Registry of Persons (BRP), resettlement applications and naturalisation applications for persons over 18 years old. This algorithm helps the registrant prevent "look-alike fraud". Sourced from the Rijksdienst voor Identiteitsgegevens (RvIG).

    Last change on 20th of August 2024, at 7:18 (CET) | Publication Standard 1.0
    Publication category
    Impactful algorithms
    Impact assessment
    Field not filled in.
    Status
    In use