Changemapp - Signalling changes in outdoor space

Municipality of Rijswijk

The algorithm compares new aerial photos with existing maps to detect changes in outdoor space. Think of new dormers, additions or other modifications to buildings. The signals are always first assessed by a staff member. Only after verification can changes be implemented in the BAG or BGT.

Last change on 20th of November 2025, at 7:30 (CET) | Publication Standard 1.0

• Publication category: Impactful algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

Field not filled in.

Begin date

2020-01

Contact information

[e-mail]

Responsible use

Goal and impact

The purpose of the algorithm is to identify changes to buildings and plots quickly and carefully. This allows the municipality to better keep the BAG and BGT records up-to-date.

The system supports employees in finding possible mutations and helps to perform work more efficiently.

The impact on residents is limited. The algorithm does not make decisions. It only signals possible changes on aerial photos. A staff member assesses each signal and determines whether an adjustment to the registration is needed.

Considerations

The use of this algorithm was chosen because manually checking all aerial photos is time-consuming and can produce errors. Automatic comparison finds more relevant changes and keeps the basic registration more reliable.

An important point of attention is that employees remain alert to possible error messages or noise in the signals. The municipality secures this through training, clear working agreements and periodic quality checks.

There is always room to override the system when the professional has reason to deviate from the signal.

Human intervention

Employees review all mutations spotted by the algorithm. They review the aerial photos, compare them with available data and decide for themselves:

• whether the mutation is correct;
• whether a change in the BAG or BGT is required;
• whether additional research or an on-site check is desirable.

No automatic decisions are made based on the algorithm. An employee always remains responsible for the final registration or follow-up action.

Risk management

The algorithm poses a limited risk to fundamental rights because:

• it does not process personal data in a targeted manner;
• identifiable individuals in aerial photographs are not the purpose of the processing;
• employees always do the final assessment.

Potential risks, such as erroneous mutations or unduly missed mutations, are mitigated by:

• human verification of all signals;
• technical validations and filters to reduce noise;
• periodic spot checks and reviews;
• logging of changes in the BAG/BGT.

Legal basis

• Addresses and Buildings (Basic Registration) Act (BAG)
• Large-Scale Topography (Basic Registration) Act (BGT)
• Environment Act
• Municipal Act
• General Data Protection Regulation (AVG), insofar as applicable to visual material

Links to legal bases

• BAG: https://wetten.overheid.nl/BWBR0022458
• BGT: https://wetten.overheid.nl/BWBR0035282
• Environment Act: https://wetten.overheid.nl/BWBR0037041

Operations

Data

The algorithm uses data needed to signal changes to buildings and plots, including:

• aerial photographs of the territory;
• existing BAG data, such as building contours;
• existing BGT data, such as topographical objects;
• metadata on photo capture data.

Data are only used for keeping basic registrations up to date and are processed according to the AVG.

Technical design

The algorithm uses image comparison to detect mutations. It works as follows:

1. New aerial photos are read in and compared with previous photos and existing BAG/BGT contours.
2. The algorithm marks spots where a difference is visible that may indicate a building change.
3. The mutations are scored and filtered to reduce noise.
4. The results are displayed in an overview to employees.
5. The employee assesses the signal and decides whether a change in registration is needed.
6. If confirmed, the change is processed in the BAG or BGT according to regular management procedures.

The system does not automatically adjust itself. Improvements in filters or working methods are made manually by the supplier or the management organisation.

External provider

Geronimo.AI